Ω re ee th mene -
- eee ries >
oes
HANDBOUND
AT THE
7
ee
-
—
-
υ
᾿
ly
f roa
j
᾽ " Ὁ
1
i
i
"1
)
i ᾿
᾿ i
υ om
I
’
i
Tt ᾿
Z i
τ
fe 5
1
a ᾿ if
” i ᾿ ἵν 7
i ᾿ ὃ i co
it i e
Ἷ ©
1
i i
᾿ : ἮΝ j
ΠΣ ;
CLASSICAL
REVIEW.
VOLUME IV.
Wondon :
DAVID NULE 270 STRAND:
1890.
iv
THE CLASSICAL REVIEW.
Me. 5.
PAGE
A Future Life as represented by the
Greek Tragedians. M. M. Danien 81 TORR > ¥ . 126
Political Allusions in the Supplices of Schmidt’s Handbook of Greek and Latin
Euripides. P. GILEs . ; 95 Synonyms. A. ὃ. WILKINS . . 128
The Plot of the Agamemnon. A. PLatr 98 | Perrin’s Edition of Homer’s Odyssey,
Construction of Clauses Expressive of Books I-IV. R. P. Kerr. « 129
Expectation in Greek. W. T. Len- Hoffmann on the Indo- Germanic Present.
DRUM ἢ ee ae UY J. H. Mouton 5129
Ancient Law ‘of Search for Stolen Notes.
Property. H.C. Marcnant... .101 Thue. v. 68. Οἱ, E. Graves 129
Notes on the Theaetetus. J. Adam. . 102 Catullus Attis 62. A. Parmer. . . 130
Greek MSS. in Italian Libraries. T. W. Propertius 11, xx.—xxv. and xxv. 19.
ALLEN .. ae es omer LOS A.C. PEARSON. . . ope glad
Tucker’s Supplices. “A. E. Housman . 105 On the Iterative Use of dv. C.
Rutherford’s Thucydides IV. R. Y. HARRISON. . . Sees ει
TYRRELL .. Be aK) Greek Version of some lines of
Couat’s Ancient Attic Comedy. R. Y. Coleridge. W.T. Lenprum .. . 131
TYRRELL . . . 112 | Obituary.
Two Editions of Andocides, Morris H. Henry S. Frieze, LL.D. M. L. D’Ooge 131
MorcGan . 114 | Archaeology.
A Translation of the Argonautica of Athena Parthenos. A. 8. Murray . 132
Apollonius Rhodius. R. C. Seaton . 116 Arion, Se WerlvpeATON (.. <2 . 134
Owen’s Edition of the Tristia. G. M. Acquisitions of British Museum. C. 8. 135
Epwarpbs . “18 Brinton’s Ethnologic Affinities of the
Hardy’s Edition of the Correspondence Ktruscans. . . 136
between Pliny and eee dnt. B: ᾿Αφροδίτη ἜΣ ΝΣ "7. Warkiss
Mayor. . . 120 Lioyp . 136
Bury’s History of the Later Roman Summaries of Periodicals 137
Empire. T. HopeKin . . . 124 | Bibliography. . . 142
No.4.
The Gameof Harpastum. Οὐ. E. Marinpin145 | Luchs’ Livy. Vol.iv. M. T. Tarnam 181
Echoed Phrases in the ee G. Ebert’s History of Later Latin Litera-
C.-M. Suira =. Os 2. sie bore, τ Ee ἢ. 1 x. pi er 1 eed
Mr. Whitelaw on 2 Cor. vi. 2—vii. 1. Edwardes’ Sardinia. Ο. Torr 181
F. H. Cuase . path ΣΕ Mule loO.| Notes!
Studies in Caesar. I. E.G. Srauer . 152 Aesch. Ag. 425. J. B. Bury 182
Horatiana. J. Gow . 154 374. 8. J. WARREN 182
Omentum. G. F. SrTrt . : . 156 Eur. Phoen. 854. 8. THELWALL 182
Hadley’s Hippolytus. R. Wurretaw . 158 An Allusion to the Cypria in Plato.
Flagg’s Iphigenia in Tauris. <A. E. W. R. Harvie . 182
Housman . . .160 | Athanasius de Incarn. E, N. Bennerr 182
Two Books on Bite ror biud ‘J.P. Posteate 162 δολιχόσκιος. W.H. Ὁ. Rouse . 183
The New Edition of Reisig’s Vorlesun- | Hasta Ferrata. F. York Powett . 183
gen. J. HE. Nixon . Set geese 163 Manil. 1. 809. Rosryson EL.is 184
Hogue’s Greek Irregular Verbs. E. | Four Versions of Lord Tennyson’s
C. Marcuanr .166 | ‘Crossing the Bar.’ . 184
Simcox’s Language of the “New Testa | Archaeology.
ment. F, ΠΕΝΡΑΙΜ, . .168 | Site of Olba. ΟΣ Smire . . 185
Delbriick’s Indogermanic Terms of Acquisitions of Brit. Mus. , . 186
κατ ΞΡ. ΒΒ: 1) Wane ye ends. 171 Dirae. A. 5. Murray aa . 187
Soltau’s Punic of Plautus. D. S. Notes on some former Articles. Ὃ. 8. 187
MaRGOLIOUTH . . ae Latin Inscriptions in the Nottingham
The New Edition of Hermann’s Greek Museum. F.Granczer .... . 187
Antiquities. 1: Wi. WYshsc. tee |09 Wiener Vorlegebliitter. C.S. . . . 188
Toeptter’s Attische Genealogie. A. Η. Studniczka’s Kyrene, W. WrotH . 188
GREENIDGE oat . 178 Summaries . “80
Drake’s De Corona οὗ Demosthenes. Bibliography . 192
(th Edit. E. C. Marcuanr .
PAGE
Rawlinson’s History of Phoenicia. Ὁ.
TABLE OF CONTENTS.
No.
PAGE
Ropinson ELuis: Some Emendations of
Manilius . . “195
J. Gow: πεν IL Src Mavoctian
Recension. . ernie LOG
E. G. SILER : Caesar Studies I. (contd.) 198
M. A. Bayrietp: On Conditional
ἘΝ 41.2. ἐπ πὴ ἐσ 200
Classen’s Thucydides II. E. Ὁ.
MarRcHANT . 2203
ἘΠ Miuller’s Thucydides ὙΠ C.F. Sar 207
L. v. Sybel’s Bee of Plato. L.
CAMPBELL. . PCPA. aol nin 209
Headlam’s Eur. Tph. in Atul!) pba:
ENGLAND .. 2210
Coray’s Notes on Hesychius. J. B. Bory 211
Two Editions of Plautus’ Menaechmi.
E. A. SONNENSCHEIN . . . 212
A Manuscript of Catullus. ᾿ΈΟΒΙΧΘΟΝ
EELS. . : anst 214
Peter’s Fasti aS Oxia: “τὰ 6. Owen 5g BUS
Diirr’s Life of Juvenal. E. G. Harpy . 216
Helmreich’s Marcellus. J. E. B. Mayor 218
Allen’s Abbreviations in Greek MSS.
EK. Maunpe THompson.. . ξεν. 1210
Treuber’s Histor y of the Lycian Ev.
FAROE Tiel τ Se et all
G. R. Scorr: Two MSS. of Persius. . 341
R. Wuireraw: On 2 Cor. 6, a 1
areply . . 248
Poppo- ‘Stah’s Thueydides: de Ν Ἢ: N.
IOWLERI τ΄. : . 249
Bayfield’s lon ἘΠ πεν ipides. oP.
WHITELAW . . Ἐπ ΘΠ]
Orelli-Mewes’ Flows: aor: pitta ie W.
T. LENDRUM. . τ DOD
Goetz’s Corpus Glas re | Iv. ἘΠῚ
NETTLESHIP . . fede a SD
Essays by Studemund’ 5 ‘pupils. KE. A.
SONNENSCHEIN . . e205
Schaefer’s Manual oft fie ἘῸΝ ΤΟΣ for
Greek Hist. R. W. Macan... . 257
Budge’s Pseudo-Callisthenes on ‘Alexan-
der. D.S. MarGouiouTH . . . 259
Goodwin’s Moods and Tenses. D. ἘΝ
Monro. . + SPA
Way’s Iliad in "English ‘Verse. Ww;
EVERETT .. . 263
Abbott and iNathesonta, Deeehhencs
E. C. MARcHANT. . S267
Stephenson’s Livy IV. M. T. TaTHAM 268
Joannides’ Conversations in Attic
Greek: J<-D,GoopEnn .. «6242121 +7268
δ.
Kubitschek’s Imp. Rom.
criptum. E.G. Harpy .
Haigh’s Attic Theatre. A.
Kelsey’s Xen. Anab. 1-IV.
Fausset’s Student’s Cicero.
tributim dis-
W. Rees
S. R. WINANS
G. E. JEANS 2
Sedlmayer’s Selections from Ovid. 58.
G. OwEN
Giildenpenning’ ‘Siieadocet., A eas
Notes.
Eur. Rhesus (passim). C. E. PALMER :
Difficulties in Juv., Sat. I. H. M.
STEPHENSON .
Juv. 8, 192. W. Τ᾿ Lenprum
Pers.15. 1334) ἘΣ 1Ὁ Morten.
ε al /
oO δεῖνα---νήδυμος---οὔτας.
χλωρηΐς. G. E. MARINDIN .
Archaeology.
Protogenes of Kaunos.
Three Books on Lat. Epigr aphy. F.
HAVERFIELD
Tho Shank andithe, Whales
The Gortyna Inscriptions.
Journals 5
Summaries of Periodicals,
Announcements of Forthcoming Books :
List of New Books .
30:
Notes.
Pindar, Nem. 9; 16.;'10; 61. W.-R:
Harpe ᾿ 3
Phues2; 495.2; 58. Ez
1 Be Bony 2
C. Torr .
C. We 9
C. SmirH 2
(: Minera
Thucso 662. ἘΠῚ Hox. &
Eur. EL 608. C. H. KEEene
Seisachtheia. W. W. μουν.
Cic. Clu. 88 167, 169.
Archaeology.
P. SANDFORD :
Pers. 8. 1, 78. P.SANDFoRD .
Juv. 8. 2, Ue, P. SANDFORD . ‘
Hor. C. 4, 2,49. A. Housman
Dares ες gius. J. H.
Omentum. cE D. DARBISHIRE .
ΤΟΡΤΟΝ
Dr. Dorpfeld on the Greek Theatre.
JANE E. HARRISON .
Dr. Dorpfeld on the Logeion. ye E.
HaicH .
Acquisitions of ‘Be iets Museen C.
SMITH
Paton ;
τρίλιθος. C. ane
An Inscription ae om “Paphos. i we ἜΝ
Aristotle Econ. 2, 2, 23.
Journals .
Summaries of Per adits
List of New Books .
C. owe
pw weer
“1.-1 -1
PAGE
221
223
227
27
. 228
228
vi
No.
PAGE
T. W. Aten: Foreign MSS. of the
Tliad (Rome) . 289
T. C. Snow: On the Brace sion: ae
Ancient Greek . Hee Δ [ΠῚ ϊ . 293
C. Harrison: Remarks on Mr. Bay-
field’s Paper . food
FLA πα δα es Mbsatotle “πη δε iv. 01299
Verrall’s Agamemnon. L. CAMPBELL . 299
Plutarchi Moralia, ἀντ ee Vol.
eA”. SElompENn. .. eo O6
Larsen’s Studia in Plut. Mor. eee A.
HOLDEN. . ἢ": . 006
The Ambrosian Palimpsest of Piawus,
EK. A. SONNENSCHEIN . : 308
Postgate’s Catullus. 8. G. OnEn 5 310
Editions of the Iliad by Cauer, δον
and Christ. W. Lear .. 5 ls)
Bergk’s Anthologia Lyrica, ed. ied :
gb AY eae als
Xenophon’s Hallonte πὶ by ΤᾺ ΠΥ
Dowdall. H. Μ. Rrynoups . woud
Zangemeister’s Orosius 314
πῆτε Studia Ambrosiana .. 2 914
Oman’s History of Greece. A. H. Cooxr 314
Greek and Latin Grammars on the
Inductive Method. T. Τὴ. GooprELu , 315
Sonnenschein’s Latin Accidence. P
CIES. Boe. π᾿ ccc ae τ π᾿ τ} 0
Notes.
F. H. Cuase and R. WHITELAW on 2
Corn Vi 2——Val.. Te 911
No
Horatiana, Keller’s Three Classes of
MSS. J.Gow... to RU Boll
Notes on Latin Poets. ἊΣ: E. Wisusaes 340 |
Latin Aorist ἀν φῶ" ἘΣ A.
KIRKPATRICK ‘ . 042
A Peculiarity in the “Escurial “MS. of
Nonius. ὟΝ. M. Linpsay . wo 46
Apparatus Criticus ad Ciceronis Tibros
De Natura Deorum. P. ScHWENKE . 347 |
Notes on Tucker’s Sa ον W.
HEADLAM. . bod D
On Some Passages in Plato’ 5 Republic |
J. ADAM . 096 ||
Some Readings in “ie fee MS. Dot |
Plato’s Republic. Lewis CAMPBELL . 358 |
2 Cor. VI. 14—VII. 1.
Xenophon’s Hellenica by Keller.
ὟΝ. Sanpay . 359
H.
A. HoLpen 360
Hartman’s Analecta Nova Xenophontea.
H. A. Houpen. 361
THE CLASSICAL REVIEW.
pe
Way’s Translation of the Iliad. T.
C. Snow .. ae eels
Pindar, Nem. X. 5, isc. 63. AW: R.
HARDIE. . Bebe 95
Pindar, Ol. Wil ΤῊΣ R. Paran εἰς Oey
Aesch. Ag. 1273. T.C. Snow . 319
Herod. We 77. KR. Proctor . Ree 3 19,
St. John XII. 3. νάρδου motixjs. E.
N. BENNETT. . . 910
2 Cor. 111,18. . τ Ween 319
Catullus 64,107. W. Everert . 319
Persius III. 39. P.Sanprorp . 319
Fulcire. J. P. PostGate J 33e02 0
Goodwin’s Moods and Tenses. E. C.
MARCHANT MMR aS Sie P07 itt ko
Zoological Notes. D’Arcy THompson
Aiba, A oe . 320
Archaeology.
Cilician Symbols. J. THEopoRE Bent 321
Scriptural Slabs from Mesopotamia.
Hi MAL We BUDGE RO ead oS) Gam ieiee2
Greek Coins at Berlin. W. Wrote . 323
Greek Coins in British Museum. H.
N. FowLEer : 3824
Acquisitions in Brigiahs Museu C. 5, 990
Inscriptions in Nottingham Museum.
F. GRANGER . . 026
Journals, &e. : . 326
Summaries of Berindicale 338
Bibliography 342
wis
Platonis Euthyphro by Adam. L.
CAMPBELL - ᾿ ΘΟ
Schwabe’s Aelius tnd ῬΑ μὴ ash EAC:
MarcHANT oOo
Werner’s St. Paul er aineusieass A.
PLUMMER. . 301
_ A Greek Legend of St. Michael by Max
Bonnet. ἊΣ ἼΟΒΕΒΤΒΟΝ.. . 368
Intermediate Greek Lexicon. F. Ἐ
ΠΙΚΉ ΒΕ, “1370
H. D. Miiller on the Tado: ΓΕ
Verb. CHR. Cookson au fll
Moulton on the Ancient Clatsical
Drama. κὰν. S. HADLEY. . . santa
Erdmann’s History οἵ Philosophy.
Pau SHOREY ὦ OUD
Rohde’s Psyche. JANE HARRISON . 376
| Kluge on the Growth of the Iliad. Ac
FAIRBANKS ; : Bile
Thompson’s Homeric Gramma ar . 378
TABLE OF CONTENTS.
PAGE
Proverbs in the Roman Comedians. E.
ἘΣ NORRIS 1.42 ee oe os 5919
Kraemer on the Astronomica οἵ
Manilius. R. Ennis . . τ 9.09
Le Jay, Inscriptions de la Cote ‘@Or.
F. HAvVERFIELD. . .- Royle)
Models and Exercises “by ier and
Bromley. J. E. INTEXONG ee te se OU
Sargent’s Greek Iambic Verse. E. op 5, 980
Hdien L. Webster on the Gothic
Guttural. E.S. SHELDON. . . . 380
Notes.
On Clauses of ce aan C.
HARRISON. . ΘΝ
Plautus, Men. ABA, 594; mroeee 314,
432,850. E.'S. πες; wae ” 381
The Birthplace of Propertius. W. Y.
SELLAR . . "ἐξα 0 Ὁ
Prof. Tarbell on 1 Soph. ‘Aj. 651. G. E.
MARINDIN. . BCS ΤΟΝ
The Latin Aorist Subjunctive “EA.
SoNNENSCHEIN . . SELIG HALOS ἢ
Mr. Gladstone on Homer and the
Assyrian Tablets. Crciu Torr. . . 399
Apparatus Criticus to Cicero’s De Na-
tura Deorum (contd.). P. SCHWENKE. 400 |
Morris’s First Book of Thucydides.
E. Οἱ MarcHant . . . 405 |
Zacher on the Scholia of Ar istophanes.
W. G. RUTHERFORD - . att aan ct AG
Sternbach’s Anthology. J. W. Mackart 408
Benoist and Thomas, Les Poésies de
@atulless ROBINSON BinISis)., - « 40)
Von Ponor’s Festus. H. Netriesuir . 412
The Vienna Corpus: Weihrich’s Specu-
lum and pence ΘΙ Priscillian. W.
SANDAY. . es eer Sy
Teuttel’s ratiees Al) _Eg S. Tompson 416
Schrader and Jevons, Ἐπ An-
tiquities. ΟἿ IGRARY i ga™ cies a ee SD
Christ’s Gorgias of Plato. Sr. Grorce
ΠΟ. ἌΣ a Ae A ee
Vil
PAG!
Tacitus, Dial. 10, 39; Seneca, Apoc.
9, 13. δῆς 15. ‘trent ἢ 5 ΘΝ
ΠΕΣ Versions of a Saying a ihe
Didaché. CC. Tayntor. . . 382
Aristoph. Plutus 885. B. Wyyne-
WILSON. . ΘΝ
The Shark and the aa fal C. Torr 382
American Philological Association 383
A correction. R. WHITELAW 384
Archaeology.
Acquisitions of British Museum. C. 8. 384
Journals 384
Summaries of Poviedicals 385
Bibliography 389
OE
Ipfelkofer on the Rhetorica ad Alexan-
drum. H. Ricwarps . Be +, 422
Ponickau de Isocratis Derioniedd lal,
CrARcn es ΠΩΣ
Godley’s Histories of “Tacitus. i F.
Burton. . 29
Hatfield on ΑΚ τ R. Wipers ‘ 494
Holtzapfel’s Beitrige. HvELyN ΑΒΒΟΤῚ 424
Notes.
κανθήλη. W. Lorine . 424
Eurip. Troades. A. C. Pearson 425
Hehoed ehrases., JieSucar. 0 A25
Thucyd. iv. 81. L. CAMPBELL 425
Jottings on Appuleis. EE. J.
CHINNOCK . Ae cements. <5 426
Obituary.
Professor Allen. C. L. Smitu . 426
Professor Sellar. Lewis CAMPBELL . 428
Archaeology.
Ancient Athens by Mrs. Verrall and
Miss Harrison. F. B. TarBett . 43
Acquisitions of the British Museum . 452
Journals ; 3 . 434
Summaries of ΤῈ σα ρας . 435
Bibliography . 438
vill
No.
PAGE
The Bodleian Facsimiles of the Hercu-
lanean Latin Papyri. W.M. Linpsay 441
Bugge and Bréal on the Latin Element
in Teutonic Mythology and ἐν ὦ
H. L. ΝΈΒΒΤΕΚΒ . 445
Gorgias, Encomium Helene. ᾽ν. R.
Paton aoe Nahe: Sr EO
Studies in Caesar. “Tite oo) τὶ τη τὰ . 448
Lucretius, V. 1350-3. Frank G. Moore 450
Notes on Cicero’s Letters to Atticus.
ΠΕ τοὺς
The Greek τι ρει in "ὩΣ: iHourth
Gospel. Jonn A. Cross . . 453
Apparatus Criticus to Cicero’s De Natura
Deorum (contd.). P. ScHwenke . . 454
Blaydes’ Edition of the Nubes. W. W.
Merrny™ . ᾿ 2457
Two Notices of Tewotts Republic of
ΡΟ. τ -
Trautwein on ‘ihe Py ologues of Εν
E. A. SONNENSCHEIN . . . 462
Papillon and Haigh’s Aeneid, Books if —
. 451
. 460
Wiles eR PAGE. : . 463
New Edition of Piderit’s ine @ratore.
A. S. WILKINS . . 466
Bieger on the Codex Pithoeanus “of
Persius. G. R.Scort . : . 407
Cowan’s Edition of Pillay: 5 1 1 Ὸ
Tracy PEcK . . 468
Thackeray’s Translations — from Pru
dentius. J. H. Lupron . . 470
THE CLASSICAL
REVIEW.
0:
PAGE
Batiffol’s Studia Patristica. A. RopErt-
SON” ane, . 472
King and Woolson: 5 Compar eine Grose
Cane EK. H. M.
Richardson’s Aeschines ‘against ἘΞ
phon. ΤΠ: Gs... «ease
Bénard, L’Esthétique d’Aristote. lal
Rrewarns . : : BE
Hoskier, A. Grea Carne oder AS
PLUMMER. . Ἐν: 1 7.9
Two Essays on Sicilian History. W.
Ruys Roperts . : ie LS
Ruggiero, Dizionario Epigraphica. F.
HAVERFIELD . . 479
Russell’s Translations into Greek and
Latin Verse. E. D.S. . 479
Notes.
Prof. Hogue and E. C. Marchant on
Greek Irregular Verbs 1179
Plato, Republic 532. P. SuHorry . . 480
The Agesias of Pindar. J. B. Bury 480
Thucyd. 1. 11. E.C. Marcnant . . 481
Arist. Eth. i. 1.. 8S. E. Wingoit . 481
Catullus, lxvi. and Ixvi. P. Srupson 481
A Correction. Ey. ABBoTT . 482
Archaeology.
Notes on Vases. H. A. Tupps. . . 482
Mirabilia. C. Torr . 483
Journals . 483
Summaries . 485
Bibliography . 486
The Classical Review
FEBRUARY 1890.
THE SIEGE OF PLATAEA.
Tue publication of Prof. Paley’s essay
‘On certain engineering difficulties in Thu-
cydides’ account of the escape from Plataea’
in vol. x. of the Journal of Philology may
be said to have been the beginning of many
evils for the Greeks and the barbarians also ;
that is, for Thucydides and his modern read-
ers. It was, for instance, the immediate
cause of the remarkable paper by Miiller-
Striibing in the Neue Jahrbiicher, 1885, p.
289, in which he proves, to the confusion of
all historians past, present, and to come, that
the work of Thucydides is neither more nor
less than a ‘military-didactic epic.’
To review in detail all that has been
written on the siege of Plataea, as described
by Thucydides, would be a work almost as
immense as the building of the siege wall ;
but it is possible, within brief limits, to
discuss a few of the difficulties raised by
Paley and those who have followed him.
And first it is worth while to state very
prominently a fact which Paley has not
brought sufficiently into the foreground.
The entire narrative of Thucydides demands
that we should think of the city of Plataea,
whatever its size may have been, as sur-
rounded at least in part by a belt of soft
earth of some depth. Without this, there
would have been no fruit-trees to cut down
and form into a barricade; without this, there
could have been no mound of earth, rising as
high as the city wall ; without this, the Pla-
taeans could not have dug amine under their
own walls, in order to remove the soil from
the mound outside. Out of this earth, no
doubt, the bricks were dug, which were used
in building the siege wall, and the spaces left
by the removal of the earth were of course
NO. XXX. VOL. IV.
the trenches by each side of the wall, as
was also the case in the long walls connecting
Megara and Nisaea.1
Among the strongest objections raised
against the truth of the account given by
Thucydides of the famous siege are (1) the
size of the siege wall: and (2) the small
number of men left behind to guard the city.
These were 480 in number.?
Of the size of the wall, however, we know
nothing. That depends entirely on the size
of the city, which is equally unknown. That
the existing ruins of the city walls, which
are about 24 miles long, are not to be taken
into calculation is obvious. Not only did
the Thebans raze the town entirely to the
ground in 425, but, when it had subsequently
been rebuilt, it was again pulled down in
374. But even if we allow a circuit of
4,000 yards for the siege wall, is the work
so impossible as Paley would have us
believe? Apparently half (or more) of
the Peloponnesian army were left behind
for the woxk; and this half would perhaps
amount to 20,000 men. ‘These, spread
over 4,000 yards, would give 50 men
for every ten yards of wall. Could not 50
men, in about three months, build ten yards
1 Whether such a belt exists now, can of course
easily be ascertained. If it does not, we may
still believe that it existed in the fifth century B.c.
Years of desolation—not to mention the débris of a
ruined city—must have wrought some change on the
surface of the district.
2 Miiller-Striibing refuses to avail himself of this
argument because the Spartans had forbidden the
taking of Plataea by storm, and the defenders had
merely to sit still! (1. ὁ. p. 804). But this fact also
has no better authority than Thucydides, and there-
fore should have no weight with M.-Str. Duncker
urges it strongly.
B
2 THE CLASSICAL REVIEW.
of such a wall as Thucydides describes, as-
suming a height of 15—20 feet? In a
very short time, between the ‘autumn 4
of 428 and ‘the beginning of winter,’ the
Athenians built a wall round Mitylene ; it
was a single wall, it is true, but it was fur-
nished with forts in some strong places, and
effectually shut up the city on the land side.
Taking into consideration the difference of
the numbers, the greater size of Mitylene,
and the shorter space of time, this wall in
Lesbos seems to imply quite as much
activity as the wall round Plataea. At
Pylos the wall was built in six days.!
On the size of the city also depends the
question whether 480 men were or were not
sufficient for the defence of Plataea. Those
who assume a city as extensive as the present
ruins (though Thucydides says it was οὐ pe
yaAnv) assert that 480 men were not enough
to guard it. But we have just as much
right, or more, to assume that 480 men were
thought enough—just enough, no doubt, for
economy’s sake, for the wall they had to
defend; in other words the city was not
larger than 480 men could defend. The
whole fighting force of Plataea in 479 B.c.
was 600 men. Unless the fighting force of
the town was altogether out of proportion
to the size of the city, the walls were not
larger than 480 men could maintain. Other-
wise the army was quite inadequate to take
any action in the field, if men were required
to guard the walls at home, and when we are
told that 480 men could not have resisted
the forces brought against them, we may
remember that the little fortress of Oenoe
successfully resisted the whole Peloponnesian
army. It may be added that in Paley’s view
the number of men left behind is a proof of
the large size of the city. He assumes a
population of 10,000—whether on the basis
of the 400 Plataeans (80 were Athenians)
now left behind, or on the basis of the 1,000
Plataeans who fought at Marathon, is not
clear. If the former, the proportion of
population to fighting men is large; if the
latter, we must remember that Plataea was
destroyed by the Persians in 480.
The plastering of the inner surface of the
the siege wall is thought to bea difficulty.
Where did all the mortar come from ? asks
Paley. Where, we might ask, did the mor-
tar come from for building the two and a half
1 The Athenians had 40. ships at Mitylene, and 1,000
hoplites under Paches: 7.6. about 9,000 men, and
they had to blockade the city by sea, while building
the wall (Thue. iii. 2, 18). The siege works at Nisaea
occupied a day and a half (Thue. iv. 69), at Delium
about three days (iv. 90).
miles of wall of which the ruins still exist
at Plataea (unless the walls were built
without mortar)? It came from the same
source as the stone of which the city walls
were built. Where there is lime, there is no
difficulty about mortar ; and where there is
lime-stone, there is no difficulty about lime.
The bricks of the siege walls were, no doubt,
exceedingly soft (Thucydides says that
Plataeans ‘pushed down the battlements’),
and needed protection on the side toward
the city—and, after all, the mortar was
perhaps little more than mud. At Pylos
there seems to have been no lack of the
material !
Another ‘crux’ is found in the fact that
when the Plataeans crossed the siege wall
in their escape they found water in the
outer ditch of the wall and none in
the inner. This, it is thought, is an over-
sight on the part of Thucydides. Hither
there was water in both ditches or in
neither. Why so? The water streaming
down Cithaeron would quickly fall into the
outer ditch, but it could not penetrate into
the inner ditch. Onthe other hand the area
within the siege wall, between it and the
city wall, would not be large enough to
collect much water. Under such circum-
stances there would always be less water in
the inner ditch than in the outer.
Once more, Paley points out that the
besiegers in their wall were really in more
danger than the besieged. They were in a
position where they could be attacked on
the outside by the Athenians and on the.
inside by the Plataeans. No doubt: they
might have been attacked in this manner,
if (1) the Athenians had not made up their
minds that they would not take the field
with an army against the Peloponnesians : (2)
and if Thebes had not been much nearer
Plataea than Athens was. As a fact they
were not attacked. The protracted siege and
final fall of Plataea made it clear to all the
Grecian world that the Athenians dare not
meet their old enemies in the field. From
this point of view (though they had other
reasons for desiring to take dire vengeance
on the Plataeans) the Thebans were quite
right, as a matter of policy, in urging the
siege upon the Peloponnesians.
In this way the ‘impossibilities’ and difficul-
ties of the siege seem to diminish on a closer
examination. But no scrutiny can help us
out of the difficulties which are raised by
the recent theories: that Thucydides is
describing an ideal or pattern siege (Miiller-
Striibing), or confounding the city wall with
the siege wall (Paley). Can we suppose that
THE CLASSICAL REVIEW. 3
Thucydides, writing as a historian of a
city on the borders of Attica, a city well
known before its ruin to many of his readers,
and occupied for a year after its fall by Mega-
rians aud others—a city too of which some
of the heroic defenders lived at Athens—suc-
ceeded in palming off on his contemporaries a
fictitious account of the siege? Or that he
failed to distinguish between the city wall
and the siege wall which surrounded it—a
mistake which hundreds could have cor-
rected? Would no ancient author have
pointed out the absurdity of his account, if
it were as absurd as we are asked to
believe? Is his history worth a moment’s
consideration, 7f he could not ascertain and
tell the truth about a siege on the borders
of Attica, in which every Athenian was
interested 2
EVELYN ABBOTT.
THE AGAMEMNON OF AESCHYLUS.
An ADDENDUM.
In a recent edition of the Agamemnon I
endeavoured to show that there is error in
the current hypothesis as to the story upon
which the drama is founded, and also conse-
quential error as to the dramatis personae
and the distribution of the parts. By over-
sight and forgetfulness I omitted a piece of
evidence which will, I believe, seem to many
more weighty than any that I have actually
adduced. I desire to bring it forward with-
out delay, and have obtained the kind per-
mission of the editor to do this in the
Classical Review.
My criticism related (1) to the story, and
(2) to the dramatis personae. It is import-
ant to bear in mind both the distinction and
the interdependence of these two matters.
The supposed story has for it the authority,
whatever that may be worth, of the Greek
hypothesis found in the MSS. The dramatis
personae and distribution of the parts, which
are given in our printed books, have not the
authority of the MSS., nor any authority
from tradition at all. They have been in-
vented in modern times, indeed quite re-
cently, and introduced in defiance of the
MSS., in order to bring the text and the
interpretation of the text, so far as might
be, into tolerable conformity with the story
supposed. This I have explained in my book.
But when the book was printed and published
I did not know, or rather had forgotten, that
the modern cast of the play, as now com-
monly printed, conflicts not only with the
MSS. of Aeschylus (which in such a matter
would be little), but with other testimony far
older and better, testimony indeed of the
very best and strongest kind which we have
on any subject connected with the ancient
theatre. If we are to be ruled in this matter
by tradition at all (which I do not assert or
necessarily admit), we should at least prefer
the tradition of the second century to the
tradition of the eleventh.
The principal representative of ancient
scholarship in relation to Greek drama is of
course Pollux. Writing in the full daylight
of Graeco-Roman learning, he is a very dif-
ferent witness from the anonymous Byzan-
tine revisers of the Codex Mediceus. It
happens that we have from Pollux a note on
the dramatis personae of the Agamemnon,
which, though it deals directly only with a
detail, presupposes and necessarily implies a
certain view of the whole play. The passage
runs as follows (Poll. iv. 109): ὅπότε μὴν ἀντὶ
τετάρτου ὑποκριτοῦ δέοι τινὰ τῶν χορευτῶν εἰπεῖν
ἐν ὠδῇ, παρασκήνιον καλεῖται τὸ πρᾶγμα, Os ἐν
᾿Αγαμέμνονι Αἰσχύλου: εἰ δὲ τέταρτος ὑποκριτής
τι παραφθέγξαιτο, τοῦτο παραχορήγημα ὀνομάζε-
Tal, καὶ πεπρᾶχθαί φασιν αὐτὸ ἐν Μέμνονι
Αἰσχύλου. ‘But whenever it was necessary
that, in the place of a fourth actor, one of
the chorus-performers should speak in lyric,
this is called a παρασκήνιον : see for example
the Agamemnon of Aeschylus. If there were
something incidentally spoken by a fourth
actor, this is termed a παραχορήγημα : and it
is said to have occurred in the MJemnon of
Aeschylus.’
The meaning of this is clear and undis-
puted. For the performance of a play there
were commonly provided, in addition to the
regular chorus, three actors trained for
spoken parts. As a very general rule this
number was the limit, and the plays were so
written that not more than three persons
(besides the choreutae), having parts to speak,
should be before the audience at the same
time. Pollux is here treating of the rare
exceptions to this rule. He divides them
into two kinds. The ordinary function of
the three ὑποκριταί was to deliver the dia-
logue. The most natural conception there-
B2
4 THE CLASSICAL REVIEW.
fore of a ‘fourth actor’ would be a person
speaking in or dinary dialogue (ev λόγῳ) in a
scene in which all the three regular ὑποκριταί
were already occupied. Of this however,
which is the case put second by Pollux, he
seems not to have known by his own obser-
vation a single instance. ‘It is said’, he
writes, ‘to have occurred in the Memnon of
Aeschylus,’ which play he had plainly not
read. His care in marking that he is here
speaking at second hand is “worthy of notice,
and enhances the authority of what he states
without such a limitation. The other, the
first-mentioned exception, is of a very pecu-
liar kind. It is where, in a scene requiring
the simultaneous presence of the three
regular actors, there is found another speaker
who, being a choreutes and speaking in lyric,
is not exactly a ‘fourth actor’, but, as Pollux
words it, ‘in the place of a fourth actor.’
For this he refers, as if the case were plain
and notorious, to the Agamemnon.
It will be seen on reflexion that there is
a little difficulty in understanding the nature
of this peculiar case. This ‘quasi-actor’,
says Pollux, is ‘one of the choreutae. Why
then, it might be asked, should not his part
be delivered by the ordinary chorus-leader 4
And why, since the choreutae for the purpose
of this rule never counted in the number of
the ‘actors’ at all, should this case be re-
garded as exceptional or noticeable in any
way? Weshall see the reason presently.
We are not here concerned with the ques-
tion how far the technical terms παρασκήνιον
and παραχορήγημα, as here used, were either
correct in themselves or generally recognized.
As Pollux uses them they are in a way cor-
relative, the chorus in the first place supply-
ing something extra to the stage, the fourth
actor in the second case being a sort of
addition to the chorus.' We however are
concerned only with the facts to which the
terms are applied.
In order to show the bearing of this testi-
mony on the question discussed in my book, I
will now set out (1) the dramatis personae and
distribution of the Agamemnon, as commonly
printed ; (2) the dramatis personae and dis-
tribution according to my edition. Those of
the MSS. it is scarcely worth while to dis-
cuss. Nobody defends or is likely to defend
them.
As commonly printed :
Dramatis personae.
A. Watchman.
1 For a comparison of the various uses of these
terms, see Mr. Haigh, The Attic Theatre, note on p.
212, by which note my attention was called to the
passage.
Chorus of Elders.
Clytaemnestra.
A Herald.
Agamemnon.
Cassandra.
Aegisthus,
These characters are distributed in the
play as follows :
1. Prologue. Watchman.
2. Parodos and
Saal. \ Chorus.
3. Bpisode 1. Clytaemnestra.
Chorus.
4, Stas. 2. Chorus.
j Herald.
5. Episode 2. Clytaemnestra.
ἰ Chorus.
6. Stas. 3. Chorus.
( Agamemnon.
7. Episode 3. - Clytaemnestra.
I Cassandra (silent).
8. Stas. 4. Chorus.
f Clytaemnestra.
9. Episode 4. Cassandra.
{ Chorus.
10. Interlude Agamemnon (behind
(1331—1371 the scenes).
Dindorf.) Chorus.
: Clytaemnestra.
[π᾿
11. Episode ᾿ 4 Aegisthus.
and Finale. mo
( Chorus.
It will be seen that there is here not the
least trace of the ‘fourth actor’ found in the
play by Pollux, Indeed it can scarcely be
said that the play absolutely requires three.
Very little ingenuity, certainly not more
than the ancients employed, as we are told,
to preserve their limitations in other places,
would be required to enable the mute
Cassandra of Episode 3 and the speaking
Cassandra of Episode 4 to be taken by differ-
ent maskers ; and except at this point two
actors, with the chorus, could easily perform
the whole.
I will now set out the arrangements as in
my recent edition.
Dramatis Personae.
A Watchman.
Chorus of Elders.
Clytaemnestra.
A Conspirator, leading the
Chorus of Conspirators.
A Herald.
THE CLASSICAL REVIEW. 5
Agamemnon.
Cassandra.
Aegisthus.
A Soldier of Aegisthus.
By these the different portions of the
play are spoken or sung as follows:
1. Prologue. Watchman.
2. Parodos and |
ines Chorus of Elders.
Clytaemnestra.
Conspirator.
Chorus of Elders.
Chorusof Conspirators
3. Episode 1.
4, Stas. 2. Chorus of Elders.
Conspirator.
: Herald.
5. Episode 2. Clytaemnestra.
{
Chorus of Elders.
6. Stas. 3. Chorus of Elders.
Agamemnon.
7. Episode 3. Clytaemnestra.
Cassandra (silent).
8. Stas. 4. Chorus of Elders.
Clytaemnestra.
9. Episode 4. Cassandra.
Chorus of Elders.
Agamemnon (behind
the scenes).
Chorus of Elders.
Clytaemnestra.
Conspirator,
Aegisthus.
Soldier.
Chorus of Elders.
Now if this was the arrangement known
- to Pollux, we can not only see at once the
application of his remark, but can explain
it with precision down to the minutest
peculiarity. The greater part of the play,
all but the last scene, can be performed by
the regular three actors. But in the last
scene there is a small fourth part, which
answers exactly to the description of the
ancient scholar. The scene consists of two
sections, (1) a dialogue, partly iambic but
chiefly lyric, conducted mainly by Clytaem-
nestra and the Coryphaeus (1371—1576),
and (2) the finale, in iambic and trochaic,
mainly conducted by Aegisthus, Clytaem-
nestra, and the Coryphaeus. These two
sections however are perfectly continuous.
There is no interval between them, and no
legitimate opportunity for an exit. But
Ὁ ΄“-}οεο---
10. Interlude.
11. Episode ὅ }
and Finale. |
from the evidence of the text it appeared
to me that in the finale one of the soldiers
accompanying Aegisthus must have spoken
twice, on each occasion one trochaic verse
(1650 and 1653), and also that in the pre-
ceding lyric dialogue, at 1522 (1521 Din-
dorf), the words,
οὔτ᾽ ἀνελεύθερον οἶμαι θάνατον
τῷδε γενέσθαι, ---
which are usually struck out as inexplicable,
must have been spoken by a partizan of
Clytaemnestra, that is to say, by the
Conspirator. I did not observe, what I
ought no doubt to have observed, that, as it
would be quite unnatural for the performer
here taking the person of the Conspirator to
leave the stage before the entrance of
Aegisthus and his troop, I had thus made,
to this small extent, a demand for a fourth
actor. But all the more striking, I think, is
the undesigned coincidence between my in-
dependent inference and the statement of
Pollux.
For observe: the various parts were com-
monly assumed to be distributed among the
actors in the order of their importance. In
the Agamemnon the protagonist would of
course play Clytaemnestra in the last scene
as throughout. The deuteragonist would
take Aegisthus. Of the two remaining
parts, the: Soldier, not the Conspirator,
would be given to the regular tritagonist,
as having two speeches to make instead of
only one, and also as requiring much more
impressive action. We should therefore
naturally hold, as Pollux and his authorities
held, that the speech of the Conspirator
(1522—1523) must be regarded as the ex-
ceptional fourth part. It is in lyric metre
(anapaests), not in the iambic of the ordinary
dialogue ; and so it is described by Pollux.
And, most remarkable of all, it is spoken
by a person whose ambiguous character,
between choreutes and actor, makes the
peculiar language of his description quite
simple and natural. The Conspirator is in
a sense ‘one of the choreutae.’ He stands
to the secondary chorus in much the same
relation in which the regular Coryphaeus
stands to the regular chorus ; and in fact in
my edition I have, upon this analogy, marked
his parts (as well as the one song of the
sub-chorus) by the sign XO. f. On the
other hand he is no member of the regular
chorus but, in the common technical sense,
a ὑποκριτής. When therefore, as at this
place, he recites anapaests in a scene other-
wise requiring the simultaneous presence of
three speakers (in addition to the regular
6 THE CLASSICAL REVIEW.
chorus), he is what Pollux calls him, ‘one of
the choreutae speaking in lyric in the place
of a fourth actor.’
I cannot but think that this absolute
agreement between an inference drawn from
MSS. of the fourteenth century and a state-
ment dating from the second not only sub-
stantiates the inference, but also strongly
fortifies the authority of our traditional
text. The makers of our MSS. had, it is
needless to say, not a notion of illustrating
the observation of Pollux. The words to
which his note refers are in the MSS. tacked,
in defiance of grammar, to the following
speech of Clytaemnestra, while in modern
texts, as I have already said, they are
desperately struck out. Yet there they
stand in the Codex Florentinus, as they must
have stood in the Aeschylus of Pollux, having
survived the copyists of more than a mil-
lennium, to illustrate and justify the true
tradition. Could there possibly be a fact
more encouraging to the study of those ma-
terials from which our knowledge of Greek
drama has been and is being built up ?
What, it may be asked, have we done
with the testimony of Pollux so long as we
have endeavoured to distribute the Aga-
memnon so as to accord with the Byzantine
story? It has been simply set aside, upon
one of those transparent pretexts which we
all employ when we are at a loss for argu-
ments. The copyist of Pollux, it is said,
introduced the reference to the Agamemnon
by error, because a few lines after came a
reference, quite differently worded, to the
Memnon. There is no one who will not
gladly be relieved of the supposed necessity
for such a hypothesis as this.
In conclusion I should like to call atten-
tion to the curious and unimpeachable
evidence of the difliculties besetting the
common distribution of the Agamemnon,
which is furnished by the version of Fitz-
gerald. In the course of accommodating
the earlier scenes of the play to a story and
cast of characters for which they were never
intended, it has been necessary, among other
things, to assign to the same speaker two
contradictory and irreconcilable speeches
(817—319 and 351—354 Dindorf). We
commentators, being compelled to preserve
the whole, have pushed through the place as
we could. But Fitzgerald did not so bind
himself ; and what does he do? He tacitly
remodels the scene, fusing, modifying, and
omitting, so that the contradiction wholly
disappears. The significance of this is the
greater, in that Mr. Fitzgerald was defending
no theory and indeed had probably not the
least idea that what he so calmly set aside
was not the genuine arrangement of Aeschy-
lus. But he saw, being free to see, that,
Aeschylus or not, it was intolerable to his
understanding: and he dealt with it ac-
cordingly.
A. W. VERRALL.
TABLE OF AFFINITY IN PLATO’S REPUBLIC.
In the Republic 459 E foll. Plato gives us
the arrangements which are to take the
place of marriage in his state. On certain
festival days men and women covertly
chosen by the guardians, though seemingly
selected by lot, are to be joined in a union
not lasting longer than the festivals them-
selves. There will not be more unions than
are sufficient, taking one thing with another,
to maintain the number of the male popula-
tion; but nothing is said at first about
prohibiting unions between near relations.
The children are to be taken from their
mothers and so brought up by the state that
relationship to particular parents shall re-
main unknown. No child will know its
parents, no parents their children. Men
are to be eligible for these regular and legal
unions between the ages of 25 and δῦ,
women between 20 and 40. When they
have passed these ages and ceased ‘ bearing
children to the state’ (τίκτειν τῇ πόλει or
γεννᾶν τῇ πόλει), they are to be at liberty to
form irregular unions, any possible offspring
of which is to be suppressed.
It is at this point that restrictions on the
ground of relationship are first mentioned.
A man, it is said, may form one of these
irregular unions with any woman he pleases
except daughter, granddaughter, mother
and grandmother: and a woman with any
man she pleases, not being son, grandson,
father or grandfather.!_ Brothers and sisters
are not mentioned in this particular sen-
1 Plato says daughter, daughter’s daughter, mother
and mother’s mother, and then again son, son’s son,
father and father’s father. But these make up
among them all grandchildren and grandparents ; ¢.g.
if a woman cannot marry her father’s father, a man
cannot marry his son’s daughter.
THE CLASSICAL REVIEW. 7
tence, where the forbidden degrees are first
given.
Hereupon Glaucon naturally asks how,
considering the arrangements made, father
and daughter, mother and son are to be
known. Socrates answers that all the
children born in the tenth or seventh month
after one of the festivals are to count as
children of all the men and women who
took part in the regular unions on that
occasion, and that relationship in the second
generation will follow accordingly. He
then proceeds to detine brother and sister,
who have not hitherto been mentioned,
as τὰ ἐν ἐκείνῳ TO χρόνῳ γεγονότα (ἔκγονα) ἐν
ᾧ αἱ μητέρες καὶ ol πατέρες αὐτῶν ἐγέννων. It
seems to me that these words are usually
mistranslated, and in any case they give rise
to great difficulty.
They are commonly taken to mean that
a man’s sister will be any woman born
about the same time as himself, that is,
within a certain time of a certain festival.
But a pupil of mine has pointed out to me,
what is certainly true, that under the
arrangements above stated children born
about the same time are exactly those who,
except in the case of twins, could not be
brothers or sisters. At the festival one
man was united to one woman and the
children born must be children of different
fathers and different mothers. If therefore
it was with a view to the prevention of real
incest that Plato defined relationships and
prohibited unions, he was not likely to
prohibit them to persons who could not be
relations and permit them to persons who
could. He would be granting full liberty
of incest while hindering an _ innocent
union.
If however we look again at the Greek,
we shall see that this was not Plato’s mean-
ing. The use of the imperfect tense ἐγέννων
and the absence of αὐτούς after it show
clearly that the words mean not ‘the time
at which their parents brought them into
the world,’ which would be ἐγέννησαν αὐτούς,
but ‘the time within which their parents
were having children,’ γεννᾶν being used in
the same sense in which it has been used
two or three times before in this and the
previous page. It refers therefore to the
whole time of life during which father
and mother were allowed, if the lot fell upon
them, to take part in the regular unions ;
and brothers and sisters will be all persons
born, roughly speaking, within thirty years
of one another, that being the period of
time during which a man might be having
children as the issue of regular unions, so
that a man and a woman born within that
period might possibly both have him for
father. This meaning is also clearly con-
veyed by a passage in the Zimaeus (18 D),
in which the arrangements of the Kepublic are
mentioned : νομιοῦσι δὲ πάντες πάντας αὐτοὺς
ὁμογενεῖς, ἀδελφὰς μὲν καὶ ἀδελφοὺς ὅσοιπερ ἂν
τῆς πρεπούσης ἐντὸς ἡλικίας γίγνωνται, τοὺς δ᾽
ἔμπροσθεν κ,τ.λ., Where ἡλικία naturally refers
to a considerable period of life, not to a few
weeks.
But if it was Plato’s intention under
ordinary circumstances to forbid all unions
between brothers and sisters thus defined,
he would thus have rendered all unions
whatever practically impossible. Under the
various conditions of age now stated a man
could not be united with any woman who
was older or younger than himself by less
than thirty years because she might be his
sister, nor with one thirty years younger
than himself because she might be his
daughter ; while a woman thirty years older
than himself might be his mother, and would
also be beyond the legal age for a regular
union.
But this cannot have been Plato’s mean-
ing. After the definitions of relationship,
he adds, in 461 EH, ὥστε ὃ νῦν δὴ ἐλέγομεν,
ἀλλήλων μὴ ἅπτεσθαι: ἀδελφοὺς δὲ καὶ ἀδελῴας
δώσει ὃ νόμος συνοικεῖν ἐὰν ὃ κλῆρος ταύτῃ
συμπίπτῃ Kat ἡ Πυθία προσαναιρῇ. I under-
stand the words ὥστε... ἅπτεσθαι to refer to
the irregular unions which were the last men-
tioned. Although: in 461C brothers and
sisters are curiously omitted from the list
of persons forbidden to form irregular
unions, we seem obliged by these words to
include them ; and Plato would seem ab-
solutely to forbid irregular unions between
persons who may possibly be near relations.
With regular unions the case is different.
‘Brothers and sisters,’ he says, ‘the law
will allow to be united, if the lot so fall,
and if the Pythian priestess also sanction
it by oracle.’ In these words brothers and
sisters seem to be distinguished from parents
and children (whose union would indeed
also be prevented by the limits of age laid
down), and the reference to the law and the
lot show that the regular unions only are
here intended. It is strange that Plato
should say ‘7 the lot so fall,’ because it
could not fall otherwise, possible brothers
and sisters being the only persons eligible
for these unions, as all other adults would
be possible parents and children. As to
the sanction of the oracle, we can hardly
suppose that it was to be obtained separately
for each particular couple after the lots had
ὃ THE CLASSICAL REVIEW.
been cast, although the order of the clauses
and the πρός in προσαναιρῇ would render this
the natural meaning. ‘The oracle would
have to sanction these unions once for all.
But perhaps Plato had not fully seen in
detail all the consequences of his own legis-
lation, and meant the ἔαν «.7.A. in its natural
sense. We may notice that, if the oracle
refused to sanction such unions, no unions
at all could take place.
No doubt there are some difficulties in
this interpretation of Plato’s arrangements,
but there can be no doubt as to the real
meaning of the words in which brothers
and sisters are defined, and Plato would
seem not to have thought out all the con-
sequences that would or might ensue.
Herevert RICHARDS.
CONJECTURAL EMENDATIONS IN THE UEDEA.
δεινὰ τυράννων λήματα Kal πως
ὀλίγ᾽ ἀρχόμενοι, πολλὰ κρατοῦντες,
χαλεπῶς ὀργὰς μεταβάλλουσιν.
τὸ γὰρ εἰθίσθαι ζῆν ἐπ᾿ ἴσοισιν
κρεῖσσον" ἐμοὶ γοῦν εἰ μὴ μεγάλως
ὀχυρῶς τ᾽ εἴη καταγηράσκειν.
τῶν γὰρ μετρίων πρῶτα μὲν εἰπεῖν
τοὔνομα νικᾷ, χρῆσθαί τε μακρῷ
λῷστα βροτοῖσιν" τὰ δ᾽ ὑπερβάλλοντ᾽
οὐδένα καιρὸν δύναται θνητοῖς,
μείζους δ᾽ ἄτας, ὅταν ὀργίσθῃ
190. δαίμων οἴκοις, ἀπέδωκεν.
Here is nothing to arrest us till we come
to the 7’ of v. 124; but this annihilates all
sense and construction in the sentence where
it occurs, and is therefore amended tu γ᾽ or δ᾽,
or, simpler still, discarded: the scribe who
inserted it supposed no doubt that he was
smoothing away an asyndeton. The words
now left possess a meaning, but it is
wrong. These verses set forth what the
Greeks were never tired of hearing—the
praise of the golden mean in disparage:
ment of a high estate; and it flatly con-
tradicts their tenour to say in the midst
of them be it mine to grow old in security if
not in grandeur, for this makes grandeur the
prime object of desire and security the
second. Mr. Th. Barthold corrects ἐπὶ μὴ
μεγάλοις, on modest means : the recent editors
accept this, so I pass to the next sentence.
χρῆσθαί re λῷστα ἴον αὐτά τε χρῆσθαι λῷστά
ἐστιν I will call by no worse name than
clumsy ; but the expression overgreatness is
tantamount to no profit for mortals 18 so
strange in itself that Nauck, Prinz, Verrall,
and now Wecklein agree to think v. 128
corrupt ; and external witness is still more
damaging. Mr. Verrall has already called
attention to an obscure scholion which I
present in this amended form: τὰ 0’ ὑπερ-
βάλλοντ᾽] αἱ δ᾽ ὑπερβολαὶ, φησὶν, ἀσθενεῖς
καὶ οὐ βέβαιοι τοῖς ἀνθρώποις τραχείᾳ (ita
ΒΟΡΙΡΒΙ, τῇ ἀρχαίᾳ MSS.) μεταβολῇ (ἰ.6. ὅταν
ὀργίσθη δαίμων ν. 129). This note, correct it
as I do or leave it as it was, is no comment
on our text: the text on which it is a com-
ment I should say with some confidence was
this:
τὰ δ᾽ ὑπερβάλλοντ᾽ ἄρρωστα βροτοῖς.
Remember that in any fairly ancient MS.
ἄρρωστα would be spelt dpwora, and that the
scribes omit iota subscript as often as they
insert it—indeed it is not recorded whether
the MSS. have Adora or Adora here: we then
see that the change of ὑπερβάλλοντ᾽ ἄρρωστα
to ὑπερβάλλοντα λῷστα is merely A for p, an
early and frequent error ; and λῷστα βροτοῖσιν
τὰ δ᾽ ὑπερβάλλοντ᾽ is a transposition in aid of
the metre. For the sense of the corrected
verse compare frag. 80 βροτοῖς τὰ μείζω τῶν
μέσων τίκτει νόσους. I take this opportunity
of saying that in Soph. £7. 1070 sq. ὅτι σφιν
ἤδη τὰ μὲν ἐκ δόμων | νοσεῖ, TA δὲ πρὸς τέκνων
x.t.X., Where the editors alter νοσεῖ to νοσεῖται
or voceve Or νοσεῖ δή OY νοσώδη OY ὀνοστά, and
then infer from these corrections of theirs
that the antistrophic οἰωνούς is a bacchius, I
suspect ἀρρωστεῖ or some such word was the
original and voce? a gloss. j
It will be convenient to consider v. 128
next. Mr. Verrall observing that one MS.
gives βροτοῖς for θνητοῖς has suggested that
δύναται. βροτοῖς is the remains of an explana-
tory supplement trimmed into measure by
the alteration θνητοῖς. I think it now ap-
pears that v. 128 is not Euripides at all, but
contains a duplicate of the scholion quoted
above explaining v. 127. The annotator can
hardly have written it as it stands, and its
original form I do not undertake to restore,
but I guess it to be a blend of two glosses :
(1) οὐδὲν δύναται βροτοῖς or θνητοῖς, an inter-
pretation of ἄρρωστα βροτοῖς : (2) καιρόν, a
supplement to ὑπερβάλλοντα conveying the
THE CLASSICAL REVIEW. 9
correct information that ὑπερβάλλειν means
ὑπερβάλλειν καιρόν, to overshoot the mark:
this phrase is found in Democritus ap. Stob.
flor. 18, 36 ὅσοι ἀπὸ. γαστρὸς Tas ἡδονὰς
ποιεῦνται ὑπερβεβληκότες TOV καιρόν and Plut.
Ages. 8 ὑπερβάλλων τὸν καιρόν, and 1 daresay
elsewhere.
The verses 125-6 as my correction leaves
them are translatable, or would be thought
so if they had been thus handed down, but
I do not defend them. It is impossible not
to wish away the two dipodiae πρῶτα μὲν
εἰπεῖν and χρῆσθαί τε μακρῷ : there would
then remain the straightforward sentence
τῶν yap μετρίων τοὔνομα νικᾷ | τὰ δ᾽ ὕὑπερ-
βάλλοντ᾽ ἄρρωστα βροτοῖς : here I do not take
ὄνομα to mean name, but I take τὸ τῶν μετρίων
ὄνομα to be a periphrasis for τὰ μέτρια, as
ὄνομ᾽ ὁμιλίας 15 for ὁμιλία in Or. 1082 and τὸ
ὄνομα τῆς σωτηρίας for τὴν σωτηρίαν in 7.71.
905. And I think a piece of external evi-
dence can be adduced to show that the in-
convenient words were not originally written
where we find them now.
These anapaests, especially the closing
verses 129-30, recall a well-known passage
of Herodotus, VII. 10, 13, ending with the
words φιλέει yap 6 θεὸς τὰ ὑπερέχοντα πάντα
κολούειν; and at vv. 125-6 Porson cited
another passage of equal celebrity from the
same historian, IIT. 80, 10, πλῆθος δὲ ἄρχον
πρῶτα μὲν οὔνομα πάντων κάλλιστον
ἔχει, ἐσονομ ίη ν᾿ δεύτερα δὲ, τῶν ὃ μούναρχος
ποιέει οὐδέν. Now if in Herodotus it is
ἰσονομία which is better than μοναρχία ‘ on
the one hand in name and on the other in
practice, while in Euripides it is μετριότης
which has this advantage over ὑπερβολή,
that in itself need not surprise us. But
when in the neighbouring sentence we find
ἰσονομία and μοναρχία compared by Euripides
also, it must I think surprise us that the
antithesis between name and practice should
not occur there instead of here ; and trust-
ing in Herodotus I place our two dipodiae
after v. 122:
ἣν Ἂς 5327 La) ΕΝ ΝΣ ς
τὸ γὰρ εἰθίσθαι ζὴν ἐπ’ ἴσοισιν,
πρῶτα μὲν εἰπεῖν, χρῆσθαί τε, μακρῷ
κρεῖσσον.
Examples οἵ πρῶτα μέν thus answered by re
are quoted by Paley from Hipp. 996, Heracl.
337-40, and Aesch. supp. 410: I give the
first, ἐπίσταμαι yap πρῶτα μὲν θεοὺς σέβειν,]
φίλοις τε χρῆσθαι μάδικεῖν πειρωμένοις. It is
perhaps worth mentioning that Herodotus a
few chapters further on, 85, 4, has τοιαῦτα
ἔχω φάρμακα, and that the similar words
τοιάδ᾽ oida φάρμακα occur in y. 718 of the
Medea.
Lastly I come to v. 130. Before Mr.
Verrall editors used to punctuate ὅταν
ὀργίσθη δαίμων, οἴκοις ἀπ. ; but as ἀποδιδόναι
means to pay and not to inflict a penalty,
this cannot be. Mr, Verrall therefore places
the comma after οἴκοις, but his translation
‘when fortune is angered with the house, that
is, with the increase of it’ shows that οἴκοις
wants a good deal of assistance to yield the
required sense. J propose then to write
μείζους δ᾽ ἄτας, ὅταν ὀργίσθῃ
δαίμων ὄγκοις, ἀπέδωκαν.
μ of! ?
te. μείζους ἄτας ἀπέδωκαν ὄγκοι ὅταν δαίμων
ὀργίσθῃ αὐτοῖς, towering fortunes pay ὦ heavier
penalty of ruin when heaven is angered with
them: ‘celsae grauiore casu decidunt turres.’
It would of course be possible, though I
should not commend it, to retain ἀπέδωκεν
with τὰ ὑπερβάλλοντα for its subject. The
sense of ὄγκος is common: a good instance is
Jrag. 506, ὦ τέκνον, ἀνθρώποισιν ἔστιν οἷς βίος]
ὃ σμικρὸς εὐκρὰς ἐγένεθ᾽, οἷς δ᾽ ὄγκος κακόν : the
plural however does not seem to be elsewhere
found outside technical writers. μείζους
ἄτας ἀπέδωκαν ὄγκοι resembles a good deal the
last sentence of Sophocles’s Antigone, with
its μεγάλοι λόγοι TOV ὑπεραύχων μεγάλας πληγὰς
ἀποτείσαντες. ὄγκος and οἶκος have been con-
founded, as was natural, in other places, for
instance at Jon 15, γαστρὸς διήνεγκ᾽ ὄγκον
(Brodaeus, οἶκον MSS.), where Cobet ap.
Badham adduces Dionys. Hal. ant. Rom.
ΠῚ. 11, ἀλλ᾽ ἡ μὲν ὑμετέρα πόλις ἀπὸ μείζονος
αὐχήματος ἀρχομένη εἰς ἐλάττονα ὄγκον (οἶκον
MS.) συνῆκται. Aesch. Ag. 961 should be
read thus: ἔστιν θάλασσα, τίς δέ vw κατα-
σβέσει; | τρέφουσα πολλῆς πορφύρας ἰσάργυρον]
κηκῖδα. παγκαίνιστον, εἱμάτων βαφάς. | ὄγκος
(Tycho Mommsen, οἶκος MSS.) δ᾽ ὑπάρχει
τῶνδε σὺν θεοῖς, ἄγαξ; | ἔχειν: πένεσθαι δ᾽ οὐκ
ἐπίσταται δόμος : here ὄγκος εἱμάτων means ἃ
great pile of raiment, as Herodotus has ὄγκος
φρυγάνων : ‘the sea abounds with purple to
dye our vesture, and of vesture for the
dyeing we have plenteous store’: no fear
then of dearth on either band. Porson’s
οἴκοις Will not serve, for ὑπάρχει τῶνδε fails to
convey the notion of abundance.
In conclusion I give vv. 122-30 consecu-
tively in order that their last state may be
compared with their first :
τὸ yap εἰθίσθαι ζῆν ex’ ἴσοισιν
πρῶτα μὲν εἰπεῖν χρῆσθαί τε μακρῷ
κρεῖσσον" ἐμοὶ γοῦν ἐπὶ μὴ μεγάλοις
ὀχυρῶς εἴη καταγηράσκειν.
τῶν γὰρ μετρίων τοὔνομα νικᾷ,
τὰ δ᾽ ὑπερβάλλοντ᾽ ἄρρωστα βροτοῖς"
μείζους δ᾽ ἄτας, ὅταν ὀργίσθη
δαίμων ὄγκοις, ἀπέδωκαν.
10 THE CLASSICAL REVIEW.
The following passages will not take so
long to examine,
24-26.
κεῖται δ᾽ ἄσιτος, σῶμ’ ὑφεῖσ᾽ ἀλγηδόσιν,
τὸν πάντα συντήκουσα δακρύοις χρόνον,
ἐπεὶ πρὸς ἀνδρὸς ober’ ἠδικημένη.
The old commentators took συντήκουσα for
συντηκομένη, a use without example. The
construction is now thought to be συντήκουσα
χρόνον, and τήκει βιοτήν v. 141 is quoted as
parallel. But parallel it is not: τήκει βιοτήν
is a mere equivalent of τήκει ἑαυτήν OY τήκεται:
αὐανῶ βίον in Soph. #7. 819 is just the same
thing as αὐανοῦμαι in Phil. 954, and when
Callimachus writes ὥμοσα σὸν βίον Catullus
translates it adiwro te ; moreover if to melt
time down meant anything at all it would
apparently mean to shorten time or make it
pass quickly. Clearly the sense wanted is
that which συντηκομένη would give, that of
1. A, 398, ἐμὲ δὲ συντήξουσι νύκτες ἡμέραι τε
δακρύοις ; and this Mr. Verrall elicits by
supplying σῶμα from the preceding verse.
Kuripides, I think, might have devised some-
thing more elegant than this ; and I would
credit him rather with writing τὸν πάντα
συντήκουσα δακρύοις χρόα, comparing 689
τί yap σὸν ὄμμα χρώς τε συντέτη χ᾽ ὅδε
and especially Hel. 1419, μή νυν ἄγαν σὸν
δάκρυσιν ἐκτήξῃης χρόα.
319—20.
γυνὴ yap ὀξύθυμος, ὡς δ᾽ αὔτως ἀνὴρ,
ῥάων φυλάσσειν ἢ σιωπηλὸς σοφός.
Mr. Verrall’s seems the only possible
account of the construction : possible I think
it is, though the position of σοφός is curious.
But what still perplexes me is the gender of
σιωπηλός and σοφός : I do not understand why
the parenthetic ὡς δ᾽ αὔτως ἀνήρ exerts this
influence, considering especially that the
speaker’s whole practical concern is with a
woman. It is strange if the poet, who
already in ὀξύθυμος and fawy had employed
two adjectives suiting masc. and fem. alike,
did not end the sentence with a third and
write σιωπηλόστομος These com-
pounds are frequent in tragedy : θρασύστομος,
κακόστομος, σεμνόστομος, χαλκόστομος, ἀθυρό-
στομος, αἰολόστομος, ἐλευθερόστομος.
339.
τί δ᾽ αὖ βιάζει κοὐκ ἀπαλλάσσει χερός ;
τί δ᾽ avis quite inappropriate and the τί δ᾽ οὖν
of one MS. is no better: Mr. Verrall pro-
poses τί οὖν ; but it seems clear that the
archetype had τί δ᾽ αὖ, which surely points to
τί δαί. Brunck and Porson, as is well known,
were for expelling this word from tragedy ;
but let us weigh the evidence. The text of
Aeschylus presents δαί once, at Prom. 933,
τί δαὶ φοβοίμην; here syntax rejects it and
demands δ᾽ ἂν in its stead: we infer then
that Aeschylus did not use it. The text of
Sophocles presents it once, at Ant. 318, τί
δαὶ ῥυθμίζεις ; here MS. testimony is worth
nothing, for if Sophocles wrote δὲ the scribe
had a metrical temptation to an error always
easy : we infer then that δαί was not used by
Sophocles. If now Euripides did not use it,
we might expect his text to offer two or
perhaps three suspicious instances ; but if
instead it offers at least half a dozen which
of themselves give no handle to suspicion at
all, the inference is obvious.
351—356. προυννέπω δέ σοι,
εἴ σ᾽ ἡ ᾽πιοῦσα λαμπὰς ὄψεται θεοῦ
καὶ παῖδας ἐντὸς τῆσδε τερμόμων χθονὸς,
θανεῖ: λέλεκται μῦθος ἀψευδὴς ὅδε.
vov δ᾽, εἰ μένειν δεῖ, pip” ἐφ᾽ ἡμέραν play’
ov yap τι δράσεις δεινὸν ὧν φόβος μ’ ἔχει.
Verse 356 makes no sense, has few defend-
ers, and is usually now sent into exile with
its innocent neighbour 355 for companion ;
but why it was inserted no one can say. I
offer this transposition and amendment :
θανεῖ: λέλεκται μῦθος ἀψευδὴς ὅδε"
μὴ γάρ τι δράσῃς, δεινὸν ὡς φόβος p’ ἔχει.
νῦν δ᾽, εἰ μένειν δεῖ, μίμν᾽ ἐφ᾽ ἡμέραν μίαν.
Sor I am horribly afraid lest you do mischief.
Compare 281 sg. δέδοικά oe... μή μοί τι
δράσῃς παῖδ᾽ ἀνήκεστον κακόν and frag. 608, 4,
φόβος πρόσεστι μὴ δράσωσί τι. If the first
letter » was lost the change of ἡ to οὐ was
not difficult : οὐ for 7 is found at 1. A. 1189,
and at Med. 695 we cannot tell whether 7
should be od or μή. It now becomes possible
to believe the hitherto incredible statement
of the scholiast that after this verse Didymus
read σιγῇ δόμους ἐσ Bao’ ἵν᾽ ἔστρωται λέχος.
381— 383.
GAN ἕν τί μοι πρόσαντες" εἰ ληφθήσομαι
δόμους ὑπερβαίνουσα καὶ τεχνωμένη,
θανοῦσα θήσω τοῖς ἐμοῖς ἐχθροῖς γέλων.
I should like some proof that we can say
δόμους ὑπερβαίνειν for ὀδὸν ὑπερβαίνειν : the
δόμων ὑπερβᾶσ᾽ of supp. 1049 will not suffice
and 15 moreover very uncertain. But how-
ever this may be, I think that izeo Ba t-
vovoa, which is precisely the σιγῇ δόμους
ἐσβᾶσ’ of v. 380, will be much more expres-
sive. Bothe has conjectured ὑπεμβαίνουσα.
734—740.
πέποιθα: ἸΤελίου δ᾽ ἐχθρός ἐστί μοι δόμος
735 Κρέων τε. τούτοις 6, ὁρκίοισι μὴ ζυγεὶς
[ἄγουσιν οὐ μεθεῖς ἂν ἐκ γαίας ἐμὲ
λόγοις δὲ συμβὰς καὶ θεῶν ἀνώμοτος,
THE CLASSICAL REVIEW. 11
Ν , 2X 3 (4
ψιλὸς γένοι᾽ ἂν κἀπικηρυκεύματα
οὐκ ἀντιθεῖο: τἀμὰ μὲν γὰρ ἀσθενῆ,
aN > 4 > \ XN / /
740 τοῖς δ᾽ ὄλβος ἐστὶ καὶ δόμος τυραννικός.
‘T trust you; but Creon and the house of
Pelias are my enemies; and against these,
if instead of a binding oath you make only
a verbal pact without attestation of gods,
you will be left defenceless and unable on
your part to retort their diplomatic mes-
sages,’ Here I have accepted Mr. Verrall’s
ψιλός for φίλος in v. 738, together with his
general view of the sense, which seems ab-
solutely necessitated by πέποιθα, v. 734, and
strongly though supertiuously confirmed by
vv. 743-4 ; then I have altered μὲν to μὴ in
v. 735 ; ejected with Badham the interpola-
tion which μὲν occasioned, v. 736, containing
as it does a barbarism if μεθεῖς is kept and
a tortured construction if it is changed to
μεθεῖ ; and amended ἂν πίθοιο in v. 739 to
ἀντιθεῖο, Which perhaps merely because it is
my own I prefer to Verrall’s ἀντισοῖο or
Leo’s and Munro’s ὀκνῶν πίθοιο. With τού-
τοις ψιλὸς γένοι᾽ ἂν καὶ οὐκ ἀντιθεῖο for τούτοις,
ψιλὸς γενόμενος, οὐκ ἂν ἀντιθεῖο, compare Soph.
Ant. 1279 8ᾳ. τὰ δ᾽ ἐν δόμοις | ἔοικας ἥκειν καὶ
τάχ᾽ ὄψεσθαι κακά.
856—859.
πόθεν θράσος ἢ φρενὸς ἢ
χειρὶ σέθεν τέκνων
καρδίᾳ τε λήψει
δεινὰν προσάγουσα τόλμαν ;
The upshot of the criticism bestowed on
this passage is that τέκνων must be re-
placed by an accusative answering θράσος.
Since μένος is not easy nor τέχναν adequate,
while σθένος, though it might be absorbed
by σέθεν, would not account for τέκνων, I do
not see what else the word can have been
but κότον, which is confounded now and
again with τόκον, and that with τέκνον.
Euripides seems to imitate Aesch. supp. 65
sqq. παιδὸς μόρον, ὡς αὐτοφόνως ὥλετο πρὸς
χειρὸς ἕθεν δυσμάτορος κότου τυχών.
1317.
a , -“
τί τάσδε κινεῖς κἀναμοχλεύεις πύλας ;
There is no more innocent-looking verse in
all Euripides. But Porson quotes these
passages : ὦ καινῶν ἐπῶν κινητὰ Kal μοχλευτά
from Ar. nub. 1397, following on an allusion
to the Aeolus of our poet ; τί τούσδε κινεῖς κἀ-
ναμοχλεύεις λόγους ; from two places in the
Christus patiens; and τί ταῦτα κινεῖς κἄνα-
μοχλεύεις ; τοῦτο δὴ τὸ τῶν τραγῳδῶν from
Heliodorus’s Aethiopica. All this celebrity
was never won by anything so simple as our
text: Mr. Verrall then rightly infers that
a strange word or a word strangely used
stood here in lieu of πύλας, and he proposes
éras. This will amply account for the notice
attracted ; but whether it will suit the verse
itself is not so sure: it does not seem to me
that the ὁπαί or perforations of a door are
things one can be said to ἀναμοχλεύειν any
more than one unlocks keyholes in English.
My own suggestion is tayas. πηγνύναι
means to make fast ; and accordingly a right
to mean anything that makes or is made
fast is the inalienable heritage of πάγη. To
show how various may be the meanings of a
verbal substantive no better examples could
be taken than words of this very family :
πάγος, frost or rock, πῆγμα, scaffold or rennet
for curdling. The liberty of a poet to set
colloquial use at naught and impose on πάγη
that meaning which specially appears in the
related verb πακτοῦν (compare Ar. Lys. 265,
μοχλοῖς δὲ Kal κλήθροισιν τὰ προπύλαια
πακτοῦν) ought not I think to be
doubted ; but we can go further. We know
that Euripides again bestowed this same
meaning in defiance of custom on another
cognate of πηγνύναι, and that Aristophanes
again laid hold on it as characteristic. In
Acharn. 479, Euripides, interrupted at home
in the writing of a tragedy, has acceded
with tolerable urbanity to the endless re-
quests with which he is pestered by Dicae-
opolis; but when it comes to σκάνδικά μοι
δὸς, μητρόθεν δεδεγμένος, that is too much,
and he ends the interview and returns to the
altitudes of tragedy with ἁνὴρ ὑβρίζει: κλῇε
πηκτὰ δωμάτων : the point of this we learn
from Pollux, x. 27, who informs us that
Euripides somewhere or other used the words
ide πακτὰ δωμάτων, frag. 991. Here too
then I suppose that Euripides made πάγας
mean the fastenings of a door; but πάγη
in common parlance meant nothing but a
net, and Use and Wont, ‘ grey nurses, loving
nothing new,’ promptly resented the innova-
tion through the lips of their champion
Aristophanes.
If this singular use of πάγη were found
in a MS., we should tranquilly record it in
our lexicons without suspicion or surprise.
Emendations, as is right and natural, are
less readily received ; but it happens that
our lexicons already contain a use of the
cognate πῆγμα equally unexampled and
equally destitute of MS. authority: [ mean
the ὅρκου πῆγμα γενναίως παγέν restored by
Auratus to Aesch. dg. 1198, where πῆγμα
has a meaning otherwise unknown, and the
MSS. have not πῆγμα but πῆμα.
A, E: Housman.
12 THE CLASSICAL REVIEW.
A FRAGMENT OF THE LOST EPISTLE TO THE CORINTHIANS.
THE passage, where it stands, is without
connection before or after: neither will it
fit in elsewhere in the Epistle: we have only
to remove it, and the continuity of vi. 13
and vii. 2 is unmistakable. It comes in the
middle of an impassioned outpouring of
personal affection, almost in the middle of a
sentence—mAatvvOnre καὶ ὑμεῖς ... χωρήσατε
ἡμᾶς, ‘ Enlarge your hearts jas I do...make
room in your hearts for me.’
Stanley suggests three hypotheses to ex-
plain the ‘dislocation of the argument.’
One, that St. Paul really wrote the passage
in this place, tacking it on to vi. 1, παρακα-
λοῦμεν μὴ εἰς κενὸν τὴν χάριν τοῦ Θεοῦ δέξασθαι
ὑμᾶς, these words being understood to refer
to contamination by heathen vices. But is
it credible that in the very midst, at the
very height, of his personal appeal to his
friends to bestow upon him a more generous
confidence, a fuller reciprocity of love, St.
Paul should have gone off into an exhorta-
tion to abstain from heathen vices? Between
πλατύνθητε καὶ ὑμεῖς and χωρήσατε ἡμᾶς!
Another hypothesis is, that at vi. 13 he was
interrupted in the composition of the letter.
That this paragraph was caused by some
passing reflection during the interval which
elapsed between the composition of what
precedes and what follows. Can it be believed
that vii. 2-4 was written after such an
interval; that St. Paul thought it worth
while and found it possible to resume the
broken thread of his appeal so exactly where
he left it, and so exactly at the same pitch
of eloquence and of feeling; that he yet
deliberately allowed this passage to break
the continuity which he was so careful to
preserve !/—Another suggestion is, that the
passage belongs to our First Epistle.
It is difficult to resist the suspicion that
we have here a fragment of a lost epistle to
Corinth, earlier than the First. In 1. v. 9
St. Paul says, ἔγραψα ὑμῖν ἐν τῇ ἐπιστολῇ μὴ
συναναμίγνυσθαι πόρνοις, referring to some
general precept which he had given; and
then he goes on to show how this is to be
limited in its application. Now in our First
Epistle he had so far spoken only of im-
morality within the Church. He had com-
manded them to punish the man who had
made the scandalous marriage, and in general
terms to cast out the old leaven. Nothing
that he had said could bear the interpre-
tation that he required them to cut them-
selves off from all intercourse with immoral
livers, whether within or without the pale of
the Church, and so by inference from all
intercourse with the heathen society in the
midst of which they lived. And ἐν τῇ ἐπι-
στολῇ (unless in a postscript) could not be
said of the present letter. Evidently, St.
Paul had written to the Corinthians before ;
and his letter had contained some such
command expressed in general terms, which
had been misunderstood and had created
perplexity ; and for the clearance of this,
among other reasons, Stephanas and the other
messengers had been sent by the Church, to
obtain from St. Paul a more explicit state-
ment of his meaning. I believe that we have
here a fragment of this earlier letter. And
the passage is precisely such as might have
created the misapprehension which St. Paul
in 1 Cor. v. corrects. ‘ Be ye not unequally
yoked together with unbelievers...Come out
from among them and be ye separate... Let
us cleanse ourselves from all filthiness of the
flesh and spirit.’ The passage was intended
to warn them against the contamination of
heathen vices. They understood it as a
command to separate themselves altogether
from heathen society. Heathen vices, St.
Paul explains in 1 Cor. y., are to be excluded
from the Church: they are to withdraw
from all intercourse, not with unbelievers but
with men of immoral lives calling themselves
Christians.
If the passage had been preserved, as it
very well might have been, as an extract
from a letter the rest of which had perished,
and preserved, as it would have been by the
Corinthians, with our two Epistles, there is
no great difficulty in supposing that the MS.
of it was mixed with the MS. of the Second
Epistle, and that where it was found it was
allowed to remain, under the impression that
it belonged to this.
R. WHITELAW.
THE CLASSICAL REVIEW. 13
NOTES ON THE PHARSALIA OF LUCAN.
Most of the following notes bear upon
the recent English commentary by Mr. C. E.
Haskins. To avoid misapprehension, it seems
well to state at the outset that his work is
thoroughly sound and serviceable on the
whole. Mistakes are few and far between.
But it is precisely in such a case that it is
worth while to point them out ; a book which
has many on every page deserves nothing but
neglect.
In regard to textual criticism, Mr. Has-
kins has, I think, just missed doing a notable
service to classical study. If he had _ be-
stowed a little more pains upon MS. readings,
he would have shown how Jitile of such
matter is necessary to a quite readable and
satisfactory edition of a classical author.
As it is, he has played into the hands of
his enemies. In the few places where he
does deal with the text, his suggestions are
very good, and they are made with com-
mendable caution (e.g. ‘Sabini’ in IX. 821).
Boox I.
76. Tellus extendere litora nolet
Excutietque fretum.
‘The land shall no more seek to advance
her shore-line nor throw off the sea.’
‘The negative is carried on to the second
clause, as is frequently the case in Lucan.’
In support of this Mr. Haskins quotes
two passages which fare quite simple and
not really like this one (II. 372-373, and
1X. 589-590). Here a negative for ‘ excu-
tiet’ has to be extracted from ‘ nolet.’
It would perhaps be rash to pronounce
this altogether impossible. But it is at
least unnecessary. The words may very
well mean, ‘The land will refuse to set its
shores as a long barrier to the waters, and
will shake off the sea’ (violently, so that it
returns, we may imagine, with overwhelming
force). Bentl. proposed to read ‘quaeret.’
291. Sic postquam fatus, et ipsi
In bellum prono tantum tamen
addidit irae.
H. says ‘ipse is often used by slaves of
their master, and by disciples of their
teacher.’ Here ‘his leader.’
‘Et ipsi in bellum prono’ is simply
μεμαῶτι καὶ αὐτῷ: Hager as he was in him-
self, of his own will, Curio’s words neverthe-
less gave his passion a fresh impulse.
577.
Horruit Alcides viso iam Dite Me-
gaeram.
This is not so simple as many passages on
which Mr. Haskins does give a note. Her-
cules had seen Dis in his own realm, yet
Megaera is terrible enough to make him
shudder.
Boox II.
103. Stat perhaps = πέπηγεν, ἐπανθεῖ, of
stiff, clotted blood.
587. Umbras nunquam flectente Syene.
‘i.e. S. is exactly under the equator, so
that the midday sun being vertically over-
head casts no shadow.’
Syene is on the northern tropic, not on
the equator. And the sun would be pre-
cisely overhead there only once in the year,
namely, at midsummer. It was the first
place, to a traveller going south, where this
ever occurred. Lucan exaggerates. In the
middle of winter, at Syene, the midday sun
would be about half-way between the zenith
and the horizon.
It may be added that, even if Syene were
on the equator, the sun would be precisely
overhead only twice in the year, namely, at
the vernal and autumnal equinoxes.
Book III.
168. Pauperiorque fuit tum primum
Caesare Roma.
(Lucan has just described how Caesar
broke into the treasury in spite of the re-
sistance of Metellus.)
H.’s note is brief: ‘Weise thinks that
this is a reference to Caesar’s former debts.’
The point is surely a much more obvious
one. To Lucan, writing under Nero,
‘“Caesare’ would mean ‘a Caesar’ or ‘the
emperor.’ Since the time when Caesar
broke into the treasury, Rome had often
been poorer than its master. The ‘fiscus’
rivalled and overshadowed the ‘aerarium’
(cf. Statius, Si/vae, 111. 3, 86-109).
Boox IV.
194. For ‘vellera’ cf. Varro Atae. fr. 21,
nubes ceu vellera lanae, the original of the
passage in Virgil which H. quotes. Varro
writes ‘cew vellera’: to Luean the word is
familiar as a name for a species of cloud.
168. Exiguo paulum distantia vallo.
14 THE CLASSICAL REVIEW.
‘Exiguo vallo’ is a descriptive ablative.
Two separate things are said about the
camps: they were very near each other, and
the rampart was low.
986-7. Cf. Hom. 11. xi. 477, ὄφρ᾽ αἷμα
Niapov καὶ yovvar’ ὀρώρῃ, ἃ passage which
Lucan may have had in mind when he
wrote.
526-8. Nam sol Ledaea tenebat
Sidera, vicino cum lux altissima
Cancro est:
Nox tum Thessalicas urgebat
parva sagittas.
‘The sun was in the constellation Gemini,
when at midday it is nearest to Cancer and
at midnight to Sagittarius.’ There must be
some mistake or misprint here, for it is
quite incredible that Mr. Haskins meant
that the sun passes through all the signs of
the zodiac in twenty-four hours! At mid-
night in summer the sun is still in Gemini,
next Cancer, Gemini and Cancer being then
under the northern horizon.
The passage is to be explained as follows.
‘The sun was in Gemini when his light is
highest, with Cancer close by’ (vicino
Canecro’ is like ‘exiguo vallo’ in 168, a de-
scriptive abl. or even an abl. absolute,
giving a distinct and subordinate fact about
his position). The sun is highest then, about
twenty-three degrees above the celestial
equator, and therefore rises earlier than at
any other period of the year (525, nec
segnis mergere ponto | tuncerat astra polus),
‘a short night sees Sagittarius low in the
southern sky’: or perhaps—though I think
this less likely—‘there is very little night
left after Sagittarius crosses the meridian,
little night pursues him.’
Weise, whose note is quoted by H., seems
to have had some vague conception of these
facts, but he puts Cancer on the wrong side
of Gemini (‘ praecedit Cancer solem ἢ).
Owing to the precession of the equinoxes
the ‘lux altissima’ would be nearer Cancer
than it is now.
643-4. Quisquis inest terris. in fessos
spiritus artus
Kgeritur ; tellusque viro luctante
laborat.
‘Egeritur: rises from it, ¢e. into his
body,’ H.
Rather ‘is drained into...’
‘Laborat: labours to assist him.’
This misses the point of the ablative. It
is the earth that suffers or is exhausted,
though he is the visible combatant.
667-9. Qui robore quamquam
Confisus Latio regis tamen un-
dique vires
Excivit Libycae gentis extrema-
que mundi
Signa suum comitata Iubam.
Mr. Haskins says, ‘“The power of the
King of the Libyan tribes,” 1.6. Juba. Gro-
tius suggests Lipyae gentes in apposition to
vires, a reading which, according to Oud. is
supported by some MSS.’
Grotius’s suggestion was a good one, and
Mr. Haskins’ view wants more defending
than he gives it. There are two objections
to taking gentis as a genitive. Would
Lucan write so awkward and formless a
construction as...regis...Libycae gentis4
And would he lengthen the -is of the gen.
before a vowel? As regards the latter
point, it may be noted that (a) on no theory
of the passage is there a strong pause after
gentis such as sometimes accounts for
lengthening of this nature; (4) in Mr. Heit-
land’s introduction (p. ci.) a list of peculiar
quantities in Lucan is given, which does not
include this or anything quite like it, the
writer adding, ‘This list is as nearly com-
plete as 1 can make it. There is very little
to note under this head, and it is just this
fact that is significant.’
670-675. Non fusior ulli
Terra fuit domino: qua sunt
longissima regna
Cardine ab occiduo vicinus
Gadibus Atlas
Terminat: a medio confinis
Syrtibus Hammon.
At qua lata iacet vasti plaga
fervida regni,
Distinet Oceanum ete.
This obscure description is best explained
by the fact that Lucan wrongly conceived
Africa as extending much further to the
east than it really does—extending far out
towards Ceylon (cf. X. 292 and H.’s note:
the Seres are near the sources of the Nile,
though even they have not seen the river’s
cradle), ‘A medio’ will then mean simply
‘in the middle’ (not ‘a medio cardine,’
supplying ‘ cardine’ from cardine ab occiduo).
Plaga fervida will mean the eastern region
of Libya, whence comes the sun (not the
south, as H. takes it). ‘ Distinet Oceanum,’
not ‘separates the ocean, i.e. extends to the
ocean in both directions, to the North and
South,’ but ‘separates...to the Hast and
West.’ This is more in keeping with the
ideas of the ancients. They speak of the
THE CLASSICAL REVIEW. 15
Ocean stream, it is true, as encircling the
whole globe, but that it les to the Εἰ. and
W. is more clearly realized and more
readily suggests itself (the sun rising thence
and setting in it again) than its northern
and southern circuit. Would Lucan speak
of the Mediterranean, which bounds Africa
on the N., as ‘Oceanus’ ἢ
Boox V.
93-94. Here and on I. 89 Mr. Haskins
quotes the prayer of Hecuba, ὦ γῆς ὄχημα
x.7.A, (where nothing is said about αἰθήρ or
ἀήρ), when the following quotation would
have been more apposite :
Ἔπῆς Ἂν Ni ε las / > 4 5327
ὁρᾶς τὸν ὑψοῦ τόνδ᾽ ἄπειρον αἰθέρα
Ν fol / ἢ 3.5.6 Ἂν > > ΄
καὶ γῆν πέριξ ἔχονθ᾽ ὑγραῖς ἐν ἀγκάλαις ;
τοῦτον νόμιζε Ζῆνα, τόνδ᾽ ἡγοῦ θεόν.
Eur. fr. inc. 935 (Nauck).
140. Ora quibus solvat nostro non in-
venit aevo.
Why should not ‘ora solvere’ here have
the sense which it has in 1. 98, oraque vatis
solvit—who deserve that he should lend
them utterance, unlock their lips, make them
his προφῆται Ἷ
159.
Nisi mergeris antris.
The use of si with the present indicative
found in Lucan and his contemporaries is
strange enough to require notice from a
commentator. Here mergeris would be the
strictly accurate construction. In 1. 533 ‘si
lussa secutus me vehis Hesperiam’ would
mean properly ‘if it is in obedience to dic-
tation that you are carrying me to Italy.’
It does mean, ‘if you obey me and carry me
to Italy’ (vehes, vexeris). L. 349 is another
instance. So IX. 212:
Et mihi si fatis aliena in iura venimus
Da talem, Fortuna, Iubam.
Here we are tempted to interpret, ‘if sub-
jection zs the gulf to which we are drifting’ ;
and similar explanations can be devised in
most of the other instances. But it is
simpler to recognize a certain looseness of
construction. V. 533 can hardly be ex-
plained away: for Caesar has not yet em-
barked.
Boox VI.
311. Nec sancto caruisset vita Catone.
This curious phrase is perhaps best ex-
plained by taking ‘vita’ to mean ‘society,’
‘the world.’ Martial sometimes uses the
word in a sense like this, as when he says of
his own writings :
Agnoscat mores vita legatque suos.
437. Quarum quidquid non creditur ars
est.
All that is most ineredible ({.6. horrible)
is their daily trade.
(H.’s note is ‘ars est] quam profitentur.
Sulpitius.’)
550. ‘Quodcunque’ is questionable. Lu-
can is usually precise in such constructions.
‘Quacunque...nudum...videt’ is proposed in
Weber’s note.
Boox VIL.
Crastina dira quies et imagine
maesta diurna
Undique funestas acies feret,
undique bellum.
Unde pares somnos populi noc-
temque beatam 4
26-29.
‘To-morrow’s night of horror haunted by
the sad image of the day’s events.’
‘Imago diurna’ suggests rather the fol-
lowing sense: ‘To-morrow a dreadful dream
will come, a waking vision of horror (οὐκ
ὄναρ ἀλλ᾽ ὕπαρ ἐχθρ ὁ νὴ), presenting to every
glance the deadly ranks of battle.’
On 1. 28 Mr. Haskins proceeds to resusci-
tate a view rightly condemned by Corte,
that ‘pares’ is a verb (somnos populi, ‘sleep
such as the people enjoy’). Corte himself
wished to read ‘ populis’ (for which he cites
several MSS.) in the sense of ‘the Pompeian
forces.’
Neither interpretation gives a sense at all
relevant to the context :
O felix si te vel sic tua Roma videret !
Lucan wishes that Pompey and the people of
Rome (utinam et idem contigisset populo
Romano, Schol.) had met in the land of
dreams (vel sic), if nowhere else, for a last
farewell. And line 28 must, I think, lead
up to this idea. ‘Would that the people
could enjoy like slumbers, a like night of
bliss!’ The genitive ‘populi’ is difficult,
but not unintelligible (O for like slumber to
be theirs, O that like slumbers were theirs !).
The dat. is the usual construction, but with
a ‘populo’ in the next line Lucan would
avoid it.
144,
Si liceat superis hominum conferre
labores.
‘Se. superum laboribus, “σχῆμα καθ᾽ ὅλον
καὶ μέρος.᾽ comparatio compendiaria.
185-7. What wonder that the com-
batants were filled with fear? Romans
10 THE CLASSICAL REVIEW.
everywhere felt that trouble was impending.
‘Populos’ means the nations or contingents
on the field of battle ‘whose last hour had
come.’ H. says ‘populos appears to refer
to Romans living abroad,’ quoting 1. 436,
which does not help him. This mistake
forces him to attach an unnatural sense to
‘lux extrema ’—‘ the last day worth living.’
198. Aethera seu totum discordi obsis-
tere caelo
Perspexitque polos.
Construe: aethera polosque obsistere per-
spexit.
202. Si cuncta perito
Augure mens hominum caeli nova
signa notasset
Spectari toto
mundo.
potuit Pharsalia
‘Cuncta mens hominum is probably equi-
valent to mens cunctorum hominum: perito
augure, under the guidance of a skilful
augur: but it is possible that “ cuncta
perito”’ should be taken together as “skilled
in all things”’.’ This is very unsatisfactory.
‘Cuncta’ must be taken with ‘signa.’ ‘If
through the eyes of a skilled augur the mind
of man could have marked all the altered
signs (constellations) of heaven, Pharsalia
might have been read in the whole sky.’
H. says ‘all the world might have known
that Pharsalia was being fought,’ forgetting
that this would have been expressed by toto
orbe (Schol. in toto caelo spectari Pharsalia
potuisset).
280. Sitque palam quas tot duxit Pom-
pelus in urbem
Curribus unius gentes non esse
triumphi.
‘Are not material for a single triumph.’ H.
This is ambiguous at least. The meaning
is, ‘cannot help him to a single triumph,’
‘to one triumph,’ as the sequel shows.
675. Sed tu quoque, coniunx,
Causa fugae voltusque tui: fatisque
negatum
Te praesente mori.
H. takes ‘-que’ disjunctively, ‘but...,’
a sense which it undoubtedly has in Lucan
sometimes. It is better to make ‘negatum’
a third subject to the predicate ‘causa,’ ‘and
the fact that Cornelia was not there to see
him fall’ (omitting the colon after ‘ tui’).
760. Decipitur quod castra rapit.
The singular decipitur is strange, when
the soldiers have just been spoken of in the
plural (757, putabunt). Should we read
‘decipitur qui castra rapit? (‘ victor’ in that
case being Caesar).
192.
H. quotes Caesar B.C. III. 99 for the
number of Pompeians killed at Pharsalia
(fifteen thousand). Perhaps ‘counts the
number of nations — nationalities—that
Pompey had led.’
Et Magni numerat populos.
794. Cf. Stat. heb. I. 702.
Adsiduam pelago non quaerere Delon.
(Not to have to look for Delos, not to look
in vain for it, since it is no longer φορητὰ
κυμάτεσσι παντοδαπαῖς τ᾽ ἀνέμων ῥιπαῖσιι L
quote this partly because Baehrens with his
usual boldness printed ‘nunc quaerere,’ ex-
plaining ‘nwne scripsi non codd.’),
Boox VIII.
18. Gravis est Magno quicumque
malorum
Testis adest.
Not ‘a danger to Magnus,’ but oppressive
to him. Magnus ‘cannot bear to meet’ any
witness of his defeat.
44, Tristes praesagia curas
Exagitant
is rather ‘besiege’ or ‘torment her sorrow-
ful heart’ than ‘forebodings excite gloom
. ? 5
anxiety.
Boox IX.
211. Scire mori sors prima viris, sed
proxima cogi.
Mr. Haskins’ translation does not make the
contrast clear. ‘Scire mori’ would sug-
gest by association ‘ consciscere sibi mortem.’
The things contrasted are ‘to find death by
one’s own hand,’ and ‘to have it thrust
upon one.’ So on 571 numquam successu
crescat honestum, his translation is vague
and unincisive (‘the right never grows
greater by success’). The meaning is ‘an
honourable cause is no more honourable
because it succeeds.’
578. Why does Mr. Haskins adopt the
cumbrous and infelicitous reading, ‘estque
...ubi.,.rather than estne...nisi... 4
579. ‘Ultra’ is rather ‘beyond these,’
‘elsewhere than in these’ (terra, pontus,
aer, caelum, virtus) than ‘beyond our-
selves.’
THE CLASSICAL REVIEW. Li
045.
Numenque secundum
Phorcus a juis.
‘The deity propitious to the waves.’ H.
Perhaps rather, ‘A God with only one
above him’—Posidon; he is Ποσειδάωνος
ὑποδμώς, like Proteus in Od. iv. Cf. IV.
110, sorte secunda, of Posidon himself, to
whom Zeus is superior.
Book X.
321. If any quotation is wanted on
‘multo murmure montis,’ Horace’s ‘non
sine montium clamore’ would give the key to
the sense most readily.
452. ‘Non timuit fatumque sibi promisit
iniquom’ means ‘he did not fear or look for
an adverse issue.’ This is the construction
which Mr. Haskins explains in his note on
J. 76. But he does not apply that explana-
tion here, and takes the passage quite
differently.
W. R. Harpie.
THE BODLEIAN MS.
Persius, Cod. Bodl. Auct. F. i. 15 (formerly
No. 2455).
Arter reading Mr. M. H. Morgan’s note in
the July number of the Classical Review, I
have examined and collated this MS., and
beg to send you the following notes on it, in
hopes room may be found for them in the
Review.
It consists of 16 ff. of vellum ; the text is
contained in 14 of the folios, 26 lines going
tothe page. The first folio merely contains
the statement of Leofric’s gift of the MS. to
his cathedral, and the last is blank.
The size of the pages is 15 in. X 10 in.
It is beautifully written, each letter being
most carefully formed. Mr. F. Madan pro-
nounces it to be written in the tenth century,
in Caroline minuscules, but showing pecu-
liarities which clearly prove it to have been
written in England. The open g’s are un-
doubtedly English, and Mr. W. H. Stevenson
has pointed out to me that Runic letters
are used as signs to mark the places to which
marginal scholia refer. These scholia, both
interlinear and marginal, are copious. They
appear to be by the same hand as the text,
and are certainly ail or nearly all of the
same date. The titles at the beginning, and
at the beginning of each Book, and the first
line of each Book, are painted in red and
green capitals of various sizes. The first
letter of each line throughout is a capital
painted in red. There are no other capitals
in the text.
It is bound up with four fragmentary ff. of
a treatise of Boethius on the metres used by
himself, and with 73 ff. containing Boethius’s
De Consolatione, the Persius coming at the
end of the volume thus formed, which is
strongly bound in white sheepskin. Mr.
NO. XXX. VOL. Iv.
OF PERSIUS, SAT. IIT.
Madan tells me the style of the binding
would of itself be enough to prove the
volume came from Exeter, and he considers
the binding certainly not later than the
fourteenth century.
Its history from the time of Leofric is
clear and uneventful. It bears, as stated
above, an inscription in Latin and Old
English, to the effect that Leofric, Bishop of
Exeter, gave it to the cathedral there for the
use of his successors. This inscription is
confirmed by an Evangeliarium, Bodl. MS.
Auct. D. ii. 16, which has at its beginning
an inventory of land recovered for the
cathedral by Leofric, and land, church fur-
niture, and books presented to it by him.
In this list occurs ‘liber boetii de consola-
tione,’ and (in another part of the list)
‘liber persii.’ These are clearly the two
MSS. now bound up together.
It appears from Mr. Macray’s Annals of
the Bodleian Library that when Bodley was
collecting books for his new library, in the
year 1602 (the very year in which it was
solemnly opened), the Dean and Chapter of
Exeter made him a munificent present of
eighty-one Latin MSS. from their Chapter
Library, where the books appear to have
been getting very damp. Five at least of
these volumes were among the gifts of
Leofric to his church. Among them was
our volume of Boethius and Persius, which
has thus been in the Bodleian from the date
of its opening.
To trace how the Persius MS. came into
Leofric’s hands is I suppose impossible.
Leofric was Bishop of Exeter from 1050,
when the see was founded, to 1072, having
been made Bishop of Crediton in 1046. He
is variously described as a Burgundian, and
as ‘Britonicus,’ which again is variously
σ
18 THE CLASSICAL REVIEW,
taken to mean a Breton and a Cornishman.
Yet Leofric is a Saxon name.
Wherever born, he was brought up and
educated in Lotharingia, whence he probably
came over to England as chaplain to King
Edward, when that monarch came and
assumed the crown 1042 (vid. F. E. Warren,
Leofric Missal).
On beginning to collate the MS. I was
struck by the continual agreement of its
readings with those of the MS. in the Gale
Collection at Trinity College, Cambridge,
which was apparently unknown to Jahn
(Jahn’s y is a different MS.), and the read-
ings of which Prof. Conington has printed
under the text in his edition.
Conington refers the Gale MS. (on Brad-
shaw’s authority) to the ninth or tenth
century.
I have therefore collated the Bodleian
MS. (Jahn’s £) throughout with this Gale
MS. (which for brevity I will call w) as de-
scribed by Conington. The similarity is
not quite so marked in some of the later
Satires as in the first, but throughout seems
very noteworthy.
f seems to have been very carefully written
(much more carefully than w) and corrected
with extreme care after it was written.
There are some points, it is true, in which
it goes all abroad.
(1) ἃ is inserted or admitted arbitrarily.
(2) wand eare treated as interchangeable ;
also often ¢ and y are confused, also ὁ and ἐδ,
single and double consonants, e.g. s and ss,
final zs and es.
(3) It is inconsistent as to assimilation of
letters, generally assimilating more than a,
e.g. 2. supplantat, w. subplantat.
(These points, and the divisions between
the words, have not been taken account of
in the following figures.)
In other respects the number of mere slips
uncorrected in 2 seems astonishingly small.
As to the agreement between β and w, 1
reckon that in 77 cases where the MSS.
differ, β and ὦ agree absolutely ; to these
may be added 12 where w shows a correction,
and 8 = was corrected. In 14 out of these
89 cases ὦ + αὶ stand alone against all the
MSS. known to Jahn (ed. 1843). In 2 more
ω + β have only one other MS. agreeing
with them.
In these cases of unique agreement, the
difference from the other MSS. is no doubt
generally slight: eg. 11. 45, ὠ.β. furtunare,
cet. fortunare; II. 57, w.8. mittent, cet.
mittunt ; IIIf. 75, w.8. munimenta, cet.
monumenta,
In 6 other cases B agrees with ὦ, but
mentions a v./. unknown to ὦ (N.B.—f
often superscribes συ. prefixed by a ‘vel’
or ‘al.’): eg. VI. 49, w is almost alone in
al, duco
reading ‘in luco.’ £ has in luco.
But the most noteworthy fact of all ap-
pears to me to be this, that in 43 cases B
appears to have had the reading of ὦ (in the
great mass of them certainly had it: in a
few an erasure leaves a little uncertainty
what was the original reading), and has
been corrected away from it. In 10 of these
w’s reading (as far as Jahn shows) is
unique: 6.9. V. 179, ὦ is alone in reading
actum; 8 has at cum; but it can be seen
that the ὁ was originally a ¢, and thetac.
I. 131, ὦ alone reads insecto; 8 has secto
with an erasure leaving room for two letters
before the s. So much for the agreement,
against which must be set the following
differences.
In 9 cases ὦ = βὶ as corrected. Most
of these are unimportant, for 6 makes an
obvious slip, and then corrects itself.
1
V. 15 is noticeable: ὦ teris, βὶ teres.
In one passage, V. 111, w appears to have
corrected away from f’s reading ;_ but
Conington is here doubtful about wo.
In 44 cases w stands absolutely alone, in
almost ali of them being obviously wrong.
In 8 cases β is similarly alone and wrong.
Where the MSS. are divided, ὦ and B
differ in 29 places. Among the more import: -
ant differences are: I. 57, ὦ. extet, β. extat ;
III. 78, w. sapio satis est, 8. satis est sapio.
It would seem I take it from these facts
that there must be some rather close con-
nexion between the two MSS. On the other
hand the differences clearly negative the
possibility of 6 being a simple copy of o.
From the number of wv.//. inserted in β, and
from the number of corrections in it, some
of them not merely corrections of slips but
alterations from one possible reading to
another one, it seems evident that the writer
and corrector of 8 had more than one MS.
before them. Is it possible that ὦ may be
one of two MSS. they had before them, and
that they sometimes used their wits on a
given passage before copying down either
reading?! I do not know enough about
MSS. to judge whether this is a very un-
likely hypothesis, but it appears β' is not
alone among MSS. of Persius in giving a
number of wvll., and I see Jahn complains
that it is almost impossible to classify the
MSS. of Persius in families, because they
are so up and down in their agreements and
differences: a fact which would seem most
THE CLASSICAL REVIEW. 19
‘easily explicable by supposing a scribe often
had more than one MS. before him.
Possibly 8 may be copied from the arche-
type of w. In any case the number of in-
stances where it agrees or did originally
agree with w, where all other MSS. appear
to differ, is striking ; unless indeed there be
a whole family of English MSS. which Jahn
has neglected.
It is certainly very curious that Prof.
Conington, who collated 7 MSS. for his
edition, 2 of them being Oxford ones of the
twelfth and fourteenth centuries respectively,
gives in his preface no hint that he knows of
the existence of this tenth century MS. in
the Bodleian.
I may note that Jahn’s information found
in the Berlin Library as to the readings of
β is very defective indeed, and not seldom
quite wrong. It would appear that now
and then the notes given Jahn by Hauthal
of the readings of Oxford MSS. refer to β.
e.g. V. 15: Jahn’s note reads ‘ teres (supr. i)
Oxon. antiqu. ap. Hauthal. terris β΄ Here
what Jahn supposes to be the readings of
two MSS. are really a correct and an utterly
incorrect account of the same MS. given
him by his two authorities.
As to the passages Mr. Morgan specially
asks about: VI. 26, 6 reads metuas, Harte.
right, Jahn defective; VI. 46, 6 reads
vels
uictis. VI. 51, 8 reads audeo inquit. VI.
69, Unge puer caules, mihi festa luce coque-
tur. Jahn right.
I shall be glad to furnish Mr. Morgan
with any other information he wants about
β᾽ 5 readings.
N.B.—It is of course possible that in some
instances my inferences from Conington’s
silence as to w’s reading—that it agrees with
his text—may prove fallacious. I shall hope
soon to get a sight of ὦ and test this.
G. R. Scort.
LEAF’S ILIAD.
The Iliad, with English Notes and Introduc-
tion, by Watrer Lear, Litt. D., late
Fellow of Trinity College, Cambridge.
Vol I. Books xiii.—xxiv. (London, Mac-
millan & Co. and New York, 1888). pp.
xvii. 505.
Dr. Lear’s second volume has followed his
first with comparatively little delay, a fact
in itself deserving of great commendation,
especially in the editing of an author re-
quiring such wide reading and patient toil.
His work continues to be of the same high
order as in the earlier volume ; there is the
same independence of thought, sound judg-
ment, and common sense. Nor has the
editor sickened of his task and grown care-
less as he struggles towards the end. With
Dr. Leaf it has been a labour of love to the
very last line. There is no trace of hasty or
ill-considered work in a single note; all has
been thought out with care. The notes are
full of information on all departments of
Homeric study, whether textual, linguistic, or
antiquarian. It is his manysidedness which
gives such a charm to Dr. Leaf’s notes.
He is not simply master of the dry bones of
scholarship, but his knowledge and interest
in Greek Archaeology and the history of
early institutions enable him to make the
dry bones live. Dr. Leaf has himself been
a genuine first-hand worker at Homer, and
often embodies in his notes fresh and valu-
able suggestions of his own, some of which
have already appeared from time to time in
the Journal of Hellenic Studies. Uence it
is that he is at home in the Homeric world,
and never, as so many have unwittingly
done, falls into the fatal mistake of thrusting
into his subject ideas derived from the later
ages of Greek life or from mediaeval
chivalry.
In a short introduction the editor dis-
courses afresh on the ‘ Origin of the Poems.’
In his former volume (p. xxiv.) he had ex-
pressed an opinion ‘that the original battle
at the ships belonging to the Μῆνις could
not be extricated from M-N-Z-O.’ Further
examination has now led him to alter
this view and, although the composition of
these books is highly complicated and doubt-
ful, he thinks that he has found a clue to
guide him to a conclusion as probable
as any of those hitherto aimed at. ‘This
clue is the presence or absence of the
wall and moat about the Greek Camp.’
These seem to have been absolutely un-
known to the original poem; any mention
of them is to be ascribed to the later hands.
Armed with this cviterion of the wall (which
c 2
20 THE CLASSICAL REVIEW.
Thucydides employed for estimating the
amount of the Greek forces when they
landed at Troy), he proceeds to decompose
the above-mentioned books into their com-
ponent elements. He thinks that the best
proof is to be found in Book If(XVI.). The
wall having been made a leading feature in
the assault upon the ships could not be
entirely passed over in the retreat ; but the
attempt to introduce it in II has in every
case led to hopeless confusion. This thesis
has been carefully and persuasively worked
out in the excellent introductions to the
several books and in the notes on special
passages. The wall is constantly found
likewise keeping bad company, being con-
tinually in connexion with passages which
betray signs of later style.
Dr. Leaf regards the Aristeia of Idomeneus
(N 136-672) as standing by itself, being
‘earlier than the invention of the wall, yet
not reconcilable with the Myjus.’ He like-
wise regards as another individual addition
the making of the Arms (last part of 3)
led up to by the very conservative skilful
additions to H, probably by the larger
additions in the last part of P, and by some
alteration of the early part of 3%. He
ascribes to a far inferior hand the large
additions to T, Y, and ®, whilst the end of
X shows a close analogy with the most
beautiful part of Z.
The best part of W and all O are due toa
poet of the highest order, who, if not the
author of the best portion of the Odyssey,
at least stands in close relations with that
poem. Professor Jebb’s view that he is the
poet of I is regarded as possible and
plausible.
Then follows an elaborate table setting
forth in five columns the separate strata of
which the //zad is composed in their (sup-
posed) chronological order. To the Mins
some 3400 lines are ascribed.
The introduction likewise gives an account
of the three Papyrus-fragments of the Jad.
We are never very likely to all agree as to
what are interpolations and what are not, but
as long as the spirit of progress exists, scholars
will not cease striving to solve the enigmas
which surround the bulk of the Ziad. So
long as those efforts are directed to ascer-
taining the relative ages of the various
strata all must heartily approve, but when
Dr. Leaf evinces a desire to reject Nestor’s
speeches on the ground of their inferiority
of style, it makes us wince. What! The
Iliad without Nestor! For if we take from
him his long didactic harangues, we may
retain his name, but his personality is gone.
Those who follow out the strictly critical
examination of the Poems have to guard
against the danger of forming an ideal of
the poet of the original kernel. They are
inclined to picture him to themselves as
incapable of writing a single line which is
not of the highest order (according to their
own notions of perfection). Any passages
then which do not attain to this standard
and do not deal directly with the subject of
the original Μῆνις, quickly come under their
ban. And yet all admit that some of the
noblest poetry in the liad is found, the
admittedly later parts, in the last six books.
The criticism of later years however is
all tending to the satisfactory conclusion that,
whether or not the Poems have undergone
change from time to time as their original
Aeolie gave place to Ionic forms, all parts of
them save a few sporadic lines can lay an
honest claim to great antiquity. Linguistic
and grammatical science as well as the
customs and institutions portrayed all point
to the same conclusion.
The introductions to the different books
are models of concise and clear summary.
Every point of importance in the structure of
the book is touched on, whilst the reader is
not wearied by laboured diffuseness which so
often mars similar attempts. The intro-
duction to 4 may serve as an excellent
sample.
Space of course forbids an elaborate ex-
amination of the matter in the notes. But
Dr. Leaf has left little opportunity for
anything save praise. He has made a
notable addition to our means for ulti-
mately solving that most important passage,
the Trial Scene (Σ 500), by taking ἀναίνετο
in its proper Homeric sense of ‘refused’
instead of the traditional ‘denied,’ and by
his very ingenious explanation, based on
the procedure in the Hwumenides (already
published in the Journal of Hellenic Studies).
On N 697 he has a good note, but we should
have liked something more explicit about
yvwros. Words denoting kinship are of
such importance for the knowledge of early
society that we cannot spare any lhght which
can be cast on them. The derivation from
γιγνώσκω seems absurd, I have long thought
that possibly the true explanation was that
γνωτός : γον (γόνος) = (κασι)γνητος, γνήσιος :
γεν (γένος). The wonder is that this has not
been suggested Jong ago. γνωτὸς is thus
simply a kinsman and in Homer is in process
of being narrowed down to mean brother, as
was the case ultimately with κασίγνητος and
the Latin frater. In N 130 I would prefer
to explain (following the indication of the
THE CLASSICAL REVIEW. 21
schol.) σάκος σάκει προθελύμνῳ as locking
shield upon the shield that formed the layer
in front, just as was done in the συνασπισμός
of later times.
There are some excellent notes on βλῆο
(N 288) and in the very difficult προκρόσσας
in Ξ. 36. By the way it may be noticed
that the ladies of the Homeric period had
no combs; at least Hera (Ξ 176) combs her
tresses with her fingers, which recalls the
Tragic periphrasis for fingers, κτένες χερῶν.
Is it more likely that the cestus of Aphrodite
(κεστὸν ἱμάντα) was a leatnern girdle orna-
mented with a pattern of punctures rather
thin embroidered ? Such a form of ornament
is, I believe, common among modern savages.
From 1. 214 it seems to have represented
the ταινία μαστῶν of later times rather than
the ζώνη. In & 373 one hardly sees why
Dr. Leaf takes τρίποδες -- tables, instead of
caldrons, the usual meaning of the word in
the Iliad. Nor do we feel quite satisfied
with his explanation of πυθμένι in 375. Can
the fact that we so often get the form τείνω
Ξε τίνω in Inscriptions lend any support to
νείσοντο, the MS. reading for νίσσοντο in &.
568 1
In conciusion let us again congratulate
Dr. Leaf on having achieved one of the
most remarkable pieces of scholarship
attempted for many years in England, and
may we express the hope that he will advance
into fresh fields of Homeric study.
Witiiam Ripeeway,
SCHUCKBURGH’S HERODOTUS.
Herodotos VI. With Introduction, Notes and
Maps. By E. 8. Suucksurcn, M.A. (Pitt
Press Series.) Cambridge: 1889. 4s. 6d.
Herodotos 1X. 1-89 ditto. 1887. 3s. 6d.
Tue Highth Book of Herodotus by the same
editor has not come to hand ; but, to judge
by the Preface to the first volume cited
above, is drawn up on the same lines as the
books under review, and no doubt with
equal care and success. The separate pub-
lication of a section of the Ninth Book was
presumably dictated by the exigencies of
education or of examination, and the editor
has apparently had in view all along the
needs of comparative beginners. The aids
afforded to the beginner are almost too
generous ; there will be little left for his
master to do but to see that he gets Mr.
Shuckburgh by heart! His comprehension
of the story in Herodotus is assisted by an
English analysis inserted in the text, almost
chapter by chapter, as well as by an his-
torical Introduction (pp. ix.—xl. in the case
of Book VI.,ix.—xxvi. in the caseof book 1X.)
An Historical and Geographical Index, con-
sisting of short articles on the proper names
occurring in the text, supplements the
Introduction, and leaves the editor free to
devote the Votes mainly to points of
grammar and scholarship. Some notes on
the text are inserted as a separate article.
The edition of Book IX. contains an Appen-
dix (after Abicht) on the Dialect. From
this account of the edition it will be obvious
that, assuming the notes and so forth to be
sound and pointed, these little volumes
viewed as school-books leave very little to be
desired, according to the English practice of
reducing oral instruction to mere discipline.
A painstaking passman too might, with one
of these volumes in hand, dispense with a
‘coach,’ and almost with that béte noire of
his—a lexicon.
It is fair to add that the editor holds up
a sound standard of historical criticism
in his introductions, by the employment of
the supplementary sources, of which good
examples are to be found in the figure of the
τρικάρηνος ὄφις and in the translation of
Pausanias’ account of the picture of Marathon
in the Poecile Stoa. The grammatical notes
appear on the whole admirable, Mr.
Shuckburgh having a quick sense for moods
and tenses. More advanced students would find
his textual notes worth consulting, e.g. the
defence of ἐξοδίῃσι in VI. 56, and of τε καὶ
ἐδόθη in LX. 80,and would beinterested in the
citations from Polybius passim : they will not
however learn much from the editor's his-
torical notes and criticisms, which at times
indeed may mislead or puzzle young and old.
To confine the instances to the notes on
Book VI.: ‘ Ephesos’ p. xxx. is an obvious
misprint for ‘ Ephoros,’ and 599 B.c. on
Ρ. 243 should obviously be 549 B.c. (over-
throw of the Medes): but 451 B.c. given as
the date of the ‘so called Peace of Kimon,’
p. 207, suggests a grave doubt as to a
possible confusion in the editor’s mind
between the trace with Sparta of ¢. 451 B.c.
and the supposed treaty with Persia, vari-
22 THE CLASSICAL REVIEW.
ously dated but never, I believe, assigned to
451 B.c., the year before Kimon’s last great
expedition to the East. It is very startling
to be told that ‘from Aegina the first Greek
commercial centre in Egypt, Naukratis, was
established,’ and not less startling to find
Megakles, the bridegroom of Agariste of
Sikyon, represented as grandson of Megakles
the opponent of Peisistratos, pp. 235, 236.
Mr. Shuckburgh is not a safe guide in regard
to the Alkmaeonidae ; he apparently thinks
that Xanthippos by descent belonged to that
family (p. xxxvi.). The notes on Athenian
constitutional matters leave something to be
desired. Olneis, p. 187, is only a misprint,
but the remarks on the Kleisthenic reforms
and on the position of Strategi and Archons
are not adequate or correct; and the notes
on the date of the Plataean alliance, p. 167,
are unsatisfactory. The first note seems
to incline to the conventional date 519 B.c.
and does not fairly state Grote’s argument
for 509 B.c. or thereabouts: the second note
treats the conventional date as doubtful and
gives practically Grote’s argument against
it. To describe the Spartan kingships as a
‘double monarchy ’ is obviously inexact, and
courts of ‘international arbitration’ is an
overstatement for the institutions of
Artaphernes in Ionia. That the Ionian
Amphiktyony was religious and had ‘ nothing
in common with a political league’ (p. xiii.)
is an aberration, which the note on p.
89 partially corrects. The antithesis is,
however, not strictly applicable. ‘Der
Schwerpunkt des Alterthums ruhte in dem
Gedanken dass politische und _ religiose
Interessen zusammenfielen.’ 'This statement
of Nissen’s is only unsatisfactory inasmuch
as it takes the conscious antithesis of
‘polities’ and ‘religion’ for granted in
antiquity.
We shali never perhaps get in English a
satisfactory method of transliterating Greek
proper names. Scholars at present have
abandoned the old Latin equivalents and
forms without venturing to the exact re-
production of the Greek originals in English
letters. The result is an extraordinary and
mongrel nomenclature. All perhaps that can
be expected of individuals is that they should
be consistent in regard to each particular
name. Mr. Shuckburgh should make up
his mind between Skythians and Scythians,
Cambyses and Kambyses, Darius and Darios
(Dareios?), Aristides and Aristeides, and so
on. Further, if he writes ‘ Herodotos’ he
ought not to write ‘Olorus.’ ΄“ Histiaios’
and ‘ Hekataeos,’ ‘Odeium’ and ‘Olympeion’
belong to different systems of transliteration :
‘Ajax’ and ‘Pollux’ come in strangely
among the Greek forms generally adopted.
In regard to translations, there is not much
to find fault with, but ὑποθερμοτέρου is not
well rendered by ‘somewhat bitter’ nor ἐξώγ-
κωτο, ἐξωγκωμένοι by “ puffed out (up).’
It was superfluous to duplicate the note
on λέγειν, p. 176. Deliune, p. 43, and
Mytileneas, p. 239, are obvious misprints ;
but on the whole, considering the great num-
ber of different types employed and of refer- ὁ
ences given, the accuracy of the printing
is highly to be commended.
R. W. M.
HOLDEN’S LIFE
Plutarch’s Life of Timoleon, with Introduc-
tion, Notes, Maps, and Lexicon by the
Rev. H. A. HoipEn. University Press,
Cambridge, 1889. Pp. Ixxxv. and 274.
6s.
Dr, HoipeEn has already edited the lives of
Themistokles, Nikias, the Gracchi, and Sulla,
inamanner which has received high commend-
ation in America, France, and Germany, as
well as in England. The companion volume
which has just been issued has all the merits
of its predecessors, exhibiting sound scholar-
ship, sobriety of judgment and the most un-
wearied industry. The original authorities
are throughout carefully compared and no
OF TIMOLEON.
later contributions to his subject seem to
have been overlooked. The thoroughness
and completeness of the work will be seen
from a mere statement of its contents.
After a pleasantly written preface, in which
high praise is bestowed upon Mr. James
Rhoades’s Z%moleon, there follows (1) the in-
troduction of fifty pages on the primary and
secondary sources of information which were
accessible to Plutarch for his biography, and
the proof, based mainly on a careful compari-
son with the language and matter of Diodorus,
that the chief authority used by the former
was Timaeus ; (2) a chronological table of
Sicilian affairs from B.c. 740 to 337; (3)
the text (pp. 1—47); (4) explanatory notes
THE CLASSICAL REVIEW. 23
(pp. 51—151); (5) critical appendix (pp.
155—179), containing (a) an account of
MSS. ; (6) a very full bibliography, and (c)
various readings; (6) indices (pp. 182—
274), (a) of matters, (ὁ) of authors cited by
Plutarch, (6) of grammatical constructions,
followed by a list of unusual words ec.
which occur in the biography, and (d) a
complete lexical index, in which five differ-
ent marks are employed to distinguish
words of special interest.
Having stated the general character of
the work, I shall endeavour, as far as I can,
to fulfil the second duty of a reviewer, when
he comes across a really good book, and
that is to add any suggestions which may be
of use in view of a second edition.
The first point I would note is as to the
arrangement: would it not be more con-
venient to give at any rate the various read-
ings under the text? It is a nuisance to
have to refer backwards and forwards from
text to notes and again from notes to read-
ings and occasionally to index. Under the
head of misprints I have noted the follow-
ing: p. lvii. n. 99 for ὑὕπερπήνεσε read ὕπερε-
πήνεσε; p. 12, 1. 5 for s.v. μή read 8.v. od; p.
76 in note on ἀνατείνειν read ἐπανάτασις and
-σεις for ἐπανάστασις and-ces; p. 219 add
reference under δυνάστης ; 274, 1. 6 for are
read ἅτε. With these may be reckoned the
shp in p. 59, note on ili. 11 μήτε προσιόντα
τοῖς κοινοῖς ἔτι ‘although he had not yet taken
part in public business,’ which of course
should be ‘though he was no longer taking
part &e.,’ referring to his retirement from
public life after his brother’s death. Here
also may be mentioned the occasional incon-
sistencies between the explanatory and criti-
cal notes, or between the notes and index.
Examples may be found on p. 83, where the
preference seems to be given to Amyot’s
rendering of ὀψόπωλιν, ‘vivandiére,’ but in
the index the right rendering is given with-
out alternative: xxvii. 1, ὃ τὸ μὲν €ros...mpods
τὰς τροπὰς συνῆπτε τὸν καιρόν is rightly trans-
lated in the index conectebat, but the only
note on the passage is a reference to xxv. 4,
where it is quoted as an instance of the
meaning accedere : xxvii. 6 παρεγγυώντων ἄγειν
is translated in the note ‘urging him to
lead them on without delay,’ but in the
index more correctly ‘to pass the word (of
command) along the line’: viii. 3 in the
text μέν is inserted with Sintenis?, but in
crit. not. p. 170 good reason is given for its
omission.
Dr. Holden tells us that ‘ the latest recen-
sion of Sintenis forms the basis of his text,’
but that he has himself exercised an indepen-
dent judgment, availing himself both of Sin-
tenis’s collation and also of that of an Italian
MS. (M°). published by C. T. Michaelis. I
think he is perhaps inclined to err in the
same direction as Sintenis by being too
‘sparing in the introduction of good conjec-
tural emendations.’ Thus in iv. 3 Blass’
βαρέως δὴ φέρων seems required instead of ὃ έ,
the sentence, though long, not being so com-
plicated as to make Plutarch oblivious of its
construction. Inv. 15 the MS. reading φωνάς
τε δεινὰς kal κατάρας ἐπ᾽’ αὐτὸν ἀρᾶσθαι seems
to need such an insertion as the προΐεσθαι of
Blass. Inix. 1 Dr. Holden keeps σπεύδοντες
ὡς TO πέλαγος διαπλέοντες ἐκομίζοντο παρὰ τὴν
Ἰταλίαν, which he explains in the note
‘making haste, as they were crossing the
open sea,’ but the synchronistic tenses
seem to imply a contradiction. How can
they be coasting along at the same time
that they are crossing the open sea? Inthe
critical note Dr. Holden mentions the sug-
gestion διαπλεύσαντες (1 suppose with the
omission of ws), and himself suggests ‘si quid
mutandum, ws διέπλευσαν, ut sensus sit post-
quam celert cursu sunt emensi pelagus.’ Hither
reading would do, but σπεύδοντες must be
taken, I think, with ἐκομίζοντο. ix. 34 (the
Corinthians thought it impossible to over-
come the Carthaginians and the force under
Hiketas) ἣ στρατηγήσοντες ἥκοιεν “ which they
were come to take the lead of.’ Butcan the
plural be used of all the Corinthians (πᾶσι
just above)? Is not Madvig’s emendation
συστρατεύσοντες necessary? xvi. 28 (the assas-
sins employed by Hiketas who mingled with
the crowd surrounding Timoleon) ἐγγυτέρω
κατὰ μικρὸν ἐπεχείρουν. Dr. Holden translates
this ‘propius accedebant ad rem aggredien-
dam,’ but I do not see how the Greek will
bear it. Surely we must read with Madvig
ἐπεχώρουν. XXXVi. 1. 32 ἐπὶ τῆς οἰκίας ἱερὸν
ἱδρυσάμενος Αὐτοματίας ἔθυεν, αὐτὴν δὲ τὴν
οἰκίαν ἱερῷ δαίμονι καθιέρωσεν. No editor seems
to have found any difficulty in the phrase
ἱερῷ δαίμονι, but the epithet is very much out
of place: Dr. Holden would hardly translate
‘the Holy God’ with Grote. Plutarch re-
ferring to the same circumstance (Jor. p.
542 E, quoted here by Dr. Holden), says that
he dedicated his house, not ἱερῷ, but ᾿Αγαθῷ,
Δαίμονι, and surely that is what we should
read here: ἱερῷ has slipped in by mistake
from the line above. Dr. Holden has a very
attractive emendation on the difficult pas-
sage:xxx. 44 τῆς δίκης αὐτοῖς ἀπολογουμένης TH
Τιμολέοντος εὐτυχίᾳ ἐπιτιθεμένης, for which he
gives ἀπολεγομένοις τὴν Τιμολέοντος εὐτυχίαν
and translates, ‘ punishment being laid upon
them when they renounced T.’s_ success.’
24 THE CLASSICAL REVIEW,
This would do admirably if spoken of the
deserters who suffered at the hands of the
Brettians the penalty due for their sacrilegious
plunder of Delphi ; only, since it was T.’s good
fortune which had so far saved them from the
punishment due to sacrilege, I should prefer
the literal ‘good fortune’: when they
deserted, they put away from them this pro-
tecting fortune. But could so strong a word
as ἀπολέγεσθαι be used of those who were act-
ing under orders from Timoleon himself (xxx.
3. 25 and 42)? Certainly in their case the
reading suggested by Bernardakis, ἀπομονουμέ-
VOLS τῆς τ εὐτυχίας, would be more appropriate.
Another emendation of Dr. Holden’s, ἐξ ἧς
᾿εἰσέπλει (for ἔπλει) τὰ ἐπιτήδεια (xviii. 1), seems
to be almost certain. On the other hand I see
no reason for the suggested omission of τῶν
πολιτῶν in xxiv. 1, ἐπιρρεόντων πανταχόθεν εἰς
αὐτὴν τῶν πολιτῶν, Where the translation is
‘by the continual stream into it’ from all
quarters.’ But why should not Plutarch
speak of ‘its citizens streaming into it from
all quarters,’ whether we understand ‘its
citizens’ of new-made citizens, or of the
original citizens who were now returning
from exile?
I go on now to the explanatory notes.
On Proem 1. 14 it would have been better to
refer to Zeller, i. p. 836%, or some other
history of philosophy, than to give Lang-
horne’s confused and erroneous note on the
εἴδωλα οἵ Democritus; and I think συμφέ-
ρηται is hardly = accidant, but refers to the
combinations of εἴδωλα which may present
themselves to us. In 1. 6 is it necessary to
understand ἐκείνων either of ‘those lives’ or
‘the lives of them’? It is a long way from
βίων, and may be more simply translated
‘those men,’ like αὐτῶν in the next sentence.
iv. 1. 5 ὃ ἵππος πληγῇ περιπεσών Means of
course ‘ being wounded,’ as in the examples
cited ; the translation ‘receiving a cut’ is
ambiguous. vi. 2 ἵνα πράττωμεν δοκιμάσαντες,
the note and index agree in the sense ‘com-
probo’: is it not rather ‘that we may act
after full examination’? the figure which
follows bears upon want of discrimination in
eating. vil. 3 while Timoleon was busy
about the expedition letters came to Corinth
from Hiketas μηνύοντα τὴν μεταβολὴν αὐτοῦ,
is it right here to say that “ αὐτοῦ is used for
the reflexive’? the clause is written from
the point of view of the Corinthians. iv. 1.
15 δεδιότες μὴ πάθοιεν οἷα καὶ πρότερον ὑπὸ τῶν
συμμάχων ἀποβαλόντες τὴν πόλιν, is it neces-
sary to follow Held in connecting ὑπό with
ἀποβαλόντες Surely it is simpler to trans-
late ‘fearing that they should receive from
their allies the same treatment as before,
when they lost their city.’ ix. 1. 17 λόγους
τοῖς πραττομένοις ὁμοίους ‘proposals as bad as
his proceedings.’ Is it not rather ‘of the
same tenor as his negotiations with the Car-
thaginians,’ 1.6. having for their object the
exclusion of Timoleon from Sicily ? 7. 1. 20
κοινωνὸν εὖ διαπεπραγμένων ἁπάντων ‘a part-
ner in all his successes.’ I think it is better
to take the gen. as absolute. x. 1.2 τί γὰρ ἂν
καὶ περαίνειν ἀπειθῶν ; ‘for what would he gain
by withholding compliance?’ rather ‘ what
could he accomplish if he refused?’ x1. 1. 1.
οἱ Καρχηδόνιοι τοῦ Τιμολέοντος ἀνηγμένου χαλε-
πῶς φέροντες ἐν τῷ κατεστρατηγῆσθαι διατριβὴν
τοῖς Ῥηγίνοις παρεῖχον, εἰ Φοίνικες ὄντες οὐκ
ἀρέσκοιντο τοῖς δι ἀπάτης πραττομένοις, Dr.
Holden takes διατριβήν with ἐν τῷ κατεστρατη-
γῆσθαι (furnished amusement in their being
outwitted); but is not the source of the
amusement their indignation at being out-
witted? I should therefore take χιφ. with
et.k. Xi. 1. 27 ἐπιπορευόμενος ‘visiting,’
rather, as in the example quoted from Plut.
Alex. c. 31 and often in Polyb., ‘going
through the ranks.’ xiv. 1. ὃ ἐθεῶντο πολλὴν
ἐν ἀσθενέσι τοῖς ἀνθρωπίνοις καὶ προδήλοις τὴν
τῶν ἀδήλων αἰτιῶν καὶ θείων δύναμιν : here the
note is “πολλήν... ἀσθενέσι. . -προδήλοις are all
predicate-adjectives.’ It would be αἀππου
to give a predicative force to προδήλοις : one
pair of attributes ἀνθρωπίνοις καὶ προδήλοις
are opposed to another pair ἀδήλων καὶ
θείων. xvi. 1. 55 οἱ πεμφθέντες κατὰ τύχην
πυθόμενοι μέλλοντα θύειν αὐτὸν ἧκον εἰς τὸ ἱερόν:
here the note is ‘“as their good fortune
would have it” to be taken with μέλλοντα
θύειν not with πυθόμενοι, as the English
translators suppose.’ Dr. Holden’s stric-
tures on the English translators are in
general richly deserved, but here I think
they are more in the right than he is. How
can 1t be said that it was ‘their good for-
tune’ which brought the assassins to the
temple at this time, when it led to one of
them being killed on the spot and the other
at any rate failing in his aim? Dr. Holden
refers to i. 1, but there is no more reason
to translate κατὰ τύχην ‘ good fortune’ there
than here, or in xxxi. 31, where it is rightly
rendered ‘as it happened.’ It has much
the same force as ἀπὸ τύχης (1. 52, below),
which is rendered ‘forte fortuna, though I
see no reason for adding that ‘it is to be
taken with σωτηρίαν. Plutarch is continu-
ally dwelling on the providential good
fortune which accompanied Timoleon. In
each particular case it is a seeming accident
(κατὰ τύχην) Which contributes to the carry-
ing out of the Divine will. xx. 1. 22,
θαυμάζοντες τῆς θαλάσσης τὴν εὐφυίαν Kat TOV
THE CLASSICAL REVIEW. 25
χώωριων τὴν κατασκευήν, ‘expressing their ad-
miration of the convenience of the sea and
the situation of the adjacent buildings.’ 1
cannot think that these were so likely to
have engaged the attention of the soldiers
who were fishing, as the ‘productiveness of
the sea’ (Blass) and the ‘construction of the
forts.’ Just below they go on to speak of
πόλιν τοσούτοις ἐξησκημένην καλοῖς, and in xxii.
we are told how Timoleon razed to the
ground the royal forts, not sparing them,
as Dion had done, διὰ τὸ κάλλος καὶ τὴν
πολυτέλειαν τῆς κατασκευῆς (here translated
‘architecture’ by Dr. Holden). xxi. 1.
20 ἀνατραπέντων : scil. ταῖς ψυχαῖς ‘losing
heart.’ But is the word used in this sense
by itself? Why not simply ‘being over-
thrown,’ as in Plut. Pomp. p. 643 Kaicap
τοῦτον ἀνέτρεψε καὶ κατέβαλεν ἢ xxv. 1. 21,
μαινομένου παρ᾽ ἡλικίαν : Dr. Holden approves
Amyot’s version ‘ plus téméraire que son age
ne portait.’ It might have been well to
support this by quoting other passages where
παρ᾽ ἡλικίαν has the same force, as Plut. Vt.
Ῥ. 181 (of the aged Fabius), ὁρῶν τὸν Φάβιον
εὐρώστως Tap ἡλικίαν διὰ τῶν μαχομένων ὠθού-
μενον. xxvii. 1. 17, λευκάσπισι : here we are
told that ‘the Sacred Band of Carthage was
armed with huge shields covered with
elephant hides’; would it not have been
more to the point to explain that shields
were painted of different colours for the
sake of distinction, and that there was a
regiment of white-shields in the Macedonian
army ? xxvii. 1. 23, συνιδὼν τὸν ποταμὸν αὐτοῖς
ταμιεύοντα τοῦ πλήθους τῶν πολεμίων ἀπολαβεῖν
ὅσοις ἐθέλοιεν αὐτοὶ μάχεσθαι : ““ὑπεαῦ the river
was controlling for them how many they
should single off from the entire host and
engage at one time.” The middle (ταμιεύεσ-
θαι) is generally used in this sense...The
infinitive ἀπολαβεῖν depends, as an indirect
object, upon ταμιεύοντα in the same way as
after verbs of giving.’ I should rather
translate ‘perceiving that the river por-
tioned out their work to them and so cut off
from the multitude of the enemy as many as
they themselves might choose to fight with.’
For the construction of συνορᾶν with inf. cf.
2 Mace. 4. 4, συνιδὼν ᾿Απολλώνιον μαίνεσθαι.
In the passages cited in the note ταμιεύεσθαι
is used because the agent portions out to
himself : here it is the river which is sup-
posed to do so to others. xxvii. 1]. 42, τῷ
πάθει παρὰ τὸν ἀγῶνα καὶ τὸν ἐνθουσιασμὸν οὕτω
διατεινάμενος : Dr. Holden translates π. τὸν
ἀγῶντα “ because of the inward struggle.’ I
should prefer to take dyéva in its ordinary
sense, joining it with ἐνθουσιασμόν by hendia-
dys, ‘raising his voice in his excitement
Ἕλληνας,
owing to the intoxication of the combat.’
xxxvii. 1. 19, ἐφ᾽ ἃς of σοφισταὶ διὰ τῶν λόγων
τῶν πανηγυρικῶν ἀεὶ παρεκάλουν πράξεις τοὺς
ἐν αὐταῖς ἀριστεύσας: here Dr.
Holden refers to Isocrates ‘whose πανηγυρι-
kos WaS an urgent appeal to the Greeks to
combine against their common enemy the
Persians’; but it could hardly be said that
Timoleon had realized these hopes. Grote
more appropriately speaks of ‘that which
rhetors like Lysias had preached in their
panegyrical harangues, that for which Plato
sighed in the epistles of his old age—the
renewal of freedom and Hellenism through-
out Sicily.’
No two minds will be agreed as to the
exact amount of notes required. I should
have thought some unnecessary, while on
the other hand there are points left unex-
plained which might either cause perplexity
to an ordinary reader or which at any rate
would justify further illustration. Such are
the account of the attempted suicide of Ma-
mercus xxxv. 1. 34 πρός τι τῶν βάθρων δρομῷ
φερόμενος συνέρρηξε τὴν κεφαλήν ; the change
of construction in xxxv. 1. 22 ἧς μὴ προσ-
άψαιτο μηδὲ κατακοσμήσειεν ; the antecedent to
ὅθεν in xxxv. 1. 7; the way in which the
mail-clad Carthaginians were impeded by
their tunics xxviii. 1. 23; the force of γοῦν,
ib. 1. 40; the asyndeton with εἶτα xxii. 1. 5 ;
the odd phrase τὸ μὲν οὖν ἔτος ἱσταμένου θέρους
εἶχεν ὥραν xxvii. 1. 1 ; ἐμβαλὼν εἰς τὸ πέλαγος
vii. 1. 17 ; the distinction between τύραννος
and δυνάστης ; above all the hopeless sentence
xxii. 1. 17 foll, where I think the difficul-
ties should at any rate have been more fully
pointed out, 6... ὑπήκουε, without any previ-
ous mention of invitation given. If emend-
ation is despaired of, it would at any
rate have been some slight solace to the
reader to know that he had the sympathy
of the editor in his just indignation at a
writer, not being an Aeschylus or a Brown-
ing, who could leave his ideas in such an un-
developed state,
One of the best features in the book is the
lexicography ; see the excellent articles on
φιλοχωρεῖν p. 51, ἀποθεωρεῖν p. 111. foll.,
dvareivew p. 76, προχειρίζεσθαι p. 259. Here
and there the classification of meanings might
be improved: thus the phrase ἀνηγόρευον ὑπὸ
κηρύκων is given as an example of the use
of ὑπὸ with active verbs = ‘owing to,’ p.
269, and ἐμβαλὼν εἰς τὸ πέλαγος 15 given
as an instance of the meaning hostiliter
ingredi.
As the index claims to be complete it may
be as well to note the following omissions :
ἄκρος Xil. 1. 44, γηράω 111. 1. 24, ἡλικία xxv. 1.
20 THE CLASSICAL REVIEW.
23, ὠδή xiv. 1. 16, ἐφέλκεσθαι xix. 1. 33; and
why is ὀστρειογραφής untranslated ?
The list of unclassical words used by
Plutarch (p. 196) needs revision: such
words as παραλαμβάνειν, παράλογος are
thoroughly classical.
These are the gleanings which I have
been able to pick up in my examination of a
book for which the thanks of all lovers of
classical literature are due both to Dr.
Holden and to the Cambridge University
Press.
JosepH B, Mayor.
ELLIS’S COMMENTARY ON CATULLUS.
A Commentary on Catullus. By Rosrnson
Ex.is, M.A., LL.D., Fellow of Trinity
College, Oxford. University Reader in
Latin Literature. Second Edition. (Cla-
rendon Press.) 16s.
A more than common interest attends the
appearance of this book. It is in the first
place the outcome of thirty years’ continuous
labour, and as such it takes its place natu-
rally in the short list of permanent classical
works which have maintained for this
generation the best traditions of English
scholarship. Further, Mr. Ellis, who it is
fair to conjecture has here closed, for the
present at least, his series of works on this
subject, is also the father of Catullian litera-
ture in this country: and it is not given to
many men both to initiate and to consum-
mate for their own age a subject so large
and worthy as the poems of the impassioned
and divinely-gifted Veronese.
In the Preface, which is new, Mr. Ellis
traces the literary history of Catullus from
the revival of letters to the present day. It
is a chapter of extreme interest, having for
its main object the vindication of the merits
of the early Italian correctors of the text—
an essay in short upon the theme that ‘ that
age was best which was the first.’ On this
subject Mr. Ellis speaks with an authority
which it would be an impertinence in us to
criticise. His remarks come as a timely
counterblast to a since published eulogy of
Sealiger, ‘who (‘by his Catullus ete.’) first
taught criticism to walk in the road in
which it should go’ and ‘reclaimed it from
the hap-hazard guesswork’ in which ‘the
Italians had been the great offenders’ (M.
Pattison, Hssays, i. p.160—2). One wonders
whether the essayist had overlooked the
preface to Scaliger’s rather scrubby little
edition (Geneva, 1577), in which ‘Deum
testem laudo, ne integrum quidem mensem
illis tribus poetis recésendis impedimus.’
This is a custom which our English editor
happily has honoured in the breach. But
when Mr. Eljis goes on to complain that
‘for the recovery of the vera manus of
Catullus disappointingly little has been
effected since the Renaissance,’ we wish to
suggest that the reason lies as much in
the poet as in his critics. Non cuivis
homint. The peculiar quality of Catullus’s
genius offers unusual difficulties to the
emendator, as a glance at the twenty-fifth
or twenty-ninth poem will show. For
Catullus never wrote save under the impulse
of vehement emotion and he has hardly left a
line that does not bear the impress of his
passionate personality : yet at the same time
he had all Wordsworth’s gift of using the
commonest words in a far from common way
and, like Wordsworth, he was content to
draw his inspiration from ‘the common
things that round us he’
fratresque unanimos anumque matrem
and to strike the ἄφαντον φῶς of poetry
from the crudest matter :—
istos qui in platea modo hue modo illue
in re praetereunt sua occupati.
Suppose the latter half of such a line as
‘haec atque illa dies atque alia atque alia’
had been mutilated beyond recognition,
which of the critics would have been Catullus
enough to supply the words we know the
poet to have written? And, yet, what
words could suggest the long procession of
‘to-morrow and to-morrow and to-morrow’
better than this simple ‘alia atque alia’? If
the critics had been more mindful of what
Mr. Ellis happily terms the ‘inimitable
spontaneity’ of Catullus, they would have
been less prone to ‘madly thrust’ on this
rare poet the coat of a bourgeois Horace, a
blatant Ovid or a knowing Martial, and
would not, on the strength of a quotation
from one or other of these writers, have
fathered upon him such readings as ‘ furta’
for ‘facta’ (23, 10), ‘ducentum’ for ‘disertus
(12, 9), or ‘vividae’ for ‘ Lydiae’ (31, 13).
One is glad to remember that a modern
THE CLASSICAL REVIEW. 27
admirer of Catullus, who, if not a critic, is a
poet, has not found the ‘ Lydian laughter’
of Sirmio ‘too learned an allusion’ in the
last quoted passage. If ‘minuta’ for
‘inimica’ (25, 12) is ‘the finest emendation
which has been made in Catullus,’ it is a
good instance of the limitations within which
he can be emended with ‘certainty.’ One
might therefore hesitate to apply the epi-
thet ‘ certain’ even to the dozen or so con-
jectures which Mr. Ellis accepts as final
(p. xiv.), and suggest that those who deserve
best of Catullus are not the ‘aucupes
syllabarum,’ but rather the scholars who,
like Schwabe, Schmidt and Mr. Ellis himself,
have laboured to expound and illustrate what
the poet certainly wrote.
Mr. Ellis’ new edition runs to five hundred
and sixteen pages as against four hundred of
the first. Its significance is not primarily
critical, because the editor’s view of the text
remains practically what it was in 1878.
He repeats his belief in the value of the Da-
tanus and only in a very few passages does it
appear that he proposes to alter the text of his
critical edition. On 65, 2 he now accepts
Munro’s ‘Catulle’ for ‘ Catullo’—a change
surely due to the syntax: in 66, 92 he
restores the MSS. ‘eflice cur iterent,’ with
‘fulgeret’ as pres. subj.: 64, 23 the frag-
mentary ‘matrum Progenies’ is adopted asa
Grecism, ‘with a nature like your mother’s’:
16, 34 Meineke’s ‘ Cieros’ for ‘Scyros’ is
now admitted, and the same word is by a
bold conjecture intoduced into 350, ‘cum in
Ciero’ for ‘incurvo’ (MSS. incivium). The
bulk of the new matter is to be found in
excursus, discussing but rarely adopting the
views of recent critics, in occasional expan-
sions of the explanatory notes, and in the
introductions to the poems, which are in
many cases rewritten. The omissions, on
the other hand, are few: eg. the excursus
on 41, seeking to establish a legal allusion
in the poem, is now given up. (May not
the point lie in ‘non est sana’ 1. 712
Insanire is sometimes used of extravagance
in money matters: cf. Ter. Phorm. 4, 3, 37,
‘a primo homo insanibat’: Pl. Mil. 755,
‘insanivisti’: Liv. vil. 2, ‘quam ab sano
initio res in hanc vix opulentis regnis tole-
rabilem insaniam venerit’: and perhaps
Theocr. 4, 4, λυσσῆν. So here: ‘ What, give
a fright like that 10,000 HS.! insanit. Give
her a straight waistcoat—and a looking-
glass.’) And in 44 the old view, that Catul-
lus caught his cold going to Sestius’s dinner
—surely a frigid jest—is now abandoned,
and the influenza is ascribed simply to the
infection of his friend’s style. We may
compare Ar. Acharn. 138: ἔπηξε τοὺς ποτα-
povs...d7 ἐνθαδὶ Θέογνις ἠγωνίζετο.
Mr. Ellis’s critical edition must almost
have been passing through the press when
the ‘ Criticisms and Elucidations’ appeared
(1878), so that he could then do no more
than report Munro’s views in brief excursus
at the end of the volume. Jt now appears
that Mr. Ellis but rarely adopts the conclu-
sions of his great critic, to whose memory he
here (pf. xi.) pays a just and generous tribute.
In a few points Munro’s unrivalled scholar-
ship has borne fruit : ex. gr. 48, 8, where the
note on ‘ hilare,’ ete., is now omitted, οἵ. 16, 14
on ‘ putidius’ : 67, 44, speraret : 68, 143, ‘nec
tamen’: 66, 31, ‘an quod’: on 67, 34 ‘the
valuable Brit. Mus. MS. a’ has now lost its
epithet, cf. M. p. v.: 29, 21, ‘a comic
formula,’ is now omitted, cf. M. p. v., (but is
the argument saved ?): but on the larger
battle fields, such as the integrity of 54, the
circumstances of 67 and the relation between
the two parts of 68, Mr. Ellis still declines
to yield. On the last point it is interesting
to note that two of the most recent editors,
as well as Dr. Magnus, have joined Mr.
Ellis’s side in regarding Allius and Malliusas
one person. Of the critics who have written
since Munro, it does not appear that any
has pushed our knowledge of Catullus far
beyond the point reached ten years ago,
with indeed one notable exception. Mr.
Ellis has already done justice in the pages
of this Review to the merits of the Prolego-
mena to the beautiful volume with which Β.
Schmidt has recently (1887) enriched the
Tauchnitz Library. One feels therefore a
little disappointed that he has not given so
prominent a place in his commentary as was
to be expected to some of Schmidt’s leading
conclusions, especially his account of the chro-
nological order of the poems. Mr, Ellis’s own
view will be found on five or six pages (xlv.—
li.) grim with figures, from which, in some
dismay, we turn to the lucid and (in our
judgment) convincing account of Schmidt,
who tells the story of the poet’s life with a
glow worthy of the romance it really was.
Schmidt also in some interesting remarks on
Catullus’s literary position (p. lxix.) finds a
special force in the phrase ‘saecli incommoda’
(14, 22) which Mr. Ellis hesitates to accept.
Yet it seems to suit the context well,
especially if the ‘malum pedem’ of this pas-
sage is intended to carry on the hint already
given in ‘impiorum’ (l. 7: cf. Mr. Ellis’s
note), Catullus meaning to bid the ‘pro-
fanum vulgus’ of poetasters to quit the holy
ground on which they have intruded. Cf.
Prop. iii. 1, 6, ‘quo pede ingressi,’ and
28 THE CLASSICAL REVIEW.
Mayor ad Juv. x. 6: ‘dextro pede concipis.’
Further, those who agree with Schmidt's
view that infaceto rure’ (22, 14) and ‘ pleni
ruris’ (36, 4) are illustrated by Cicero’s
remark (Or. 161) on the elision of the final s
as ‘ subrusticum,’ will feel surprised that the
editor still reads ‘ beatu ’s’ (23, 37: οἵ. his
note on ‘dabi’ 116, 8).
Of Mr. Ellis’s many-sided erudition it is
now unnecessary to speak: it has been long
admitted that in this respect Catullus has
been as fortunate in his English editor as
Lucretius and Juvenal have been in theirs.
We are not less struck by the judicial tone
which dominates thiscommentary. Here are
no traces of special pleading or parti pris.
Both sides of every question are always
stated with admirable candour and lucidity,
and if the reader is ever tempted to dissent
from Mr. Ellis’s conclusions, he does so on
evidence supplied by the editor himself. To
exemplify Mr. Ellis’s method we may cite an
instance which is new to this edition. In
17, 32 the MSS. agree in ‘chinea’—a local
epithet, of which no trace seems to survive
in modern Brescia. A Venetian writer,
Zanchi (1531) quoted the verse with the
reading ‘Cygneae, from ‘nonnulla haud
contemnendae vetustatis exemplaria,’ ex-
plaining it of the arx of Brixia called after
its founder Cydnus, ‘quem Graeci Cycnum
vocant’: and it appears from later writers
—Rossi (1695), Gagliardi (1718)—that a
hill near Brescia was called ‘ Cigno’ as early
as the seventeenth century. Hence it is no
wonder that Vossius and more than one
modern editor have adopted the reading of
Zanchi. But Mr. Ellis, who acts on the
motto vade καὶ μέμνασ᾽ ἀπιστεῖν, has pursued
the question for himself through a mass of
recondite Italian literature and shows (1.)
that Zanchi, the earliest authority for the
name, does not vouch for the existence of a
hill ‘ Cigno’ in his own day: (ii.) that it is
unknown to his younger contemporary
Bagatto (1566), an antiquary and a Brescia
man, or to the earlier Capriolo (1500), whose
‘Chronica de rebus Brixianorum’ makes no
reference to ‘ Cigno,’ but several to ‘ chinnea.’
He therefore adheres to the MSS. reading,
drawing the interesting conclusion that
Zanchi’s learned support of his reading,
perhaps his own conjecture, supplanted
the real name, and instead of the reading
being due to an actual hill, the seventeenth
century Cigno is due to the reading. Simi-
larly on 56 Mr. Ellis dissents from the view
of all the commentators that the Cato there
addressed is the poetaster alluded to by
Ovid in terms that might seem to justify the
tone of Catullus’s repellent verses, and identi-
fies him with the Stoic statesman M. Porcius
Cato. Τὸ defend this position he has to put
together a chain of proof—lengthened and
strengthened in the new edition—of which the
acumen and research might compel assent,
did not this view rest on @ priori evidence
still stronger. For surely Mr. Ellis shows
true feeling for the manner of Catullus,
when he urges that in the ordinary view the
poem lacks an adequate motive. Coarse
though Catullus can be, coarseness with him
is not, as it often is with Martial, the main
object. It is not the mark at which he
aims, but rather the feather that fledges his
bolt, be it for hate or scorn or even for
pathos, as in the fifty-eighth poem, of which
‘the pity of it’ is surely the prevailing note.
There was possibly something in the
events of the day, now lost to us, which
gave special point to these verses: mean-
while we may be content to regard them,
with Mr. Ellis, as an attempt to ‘score off’
a ‘rigidus Cato.’ We could have wished
that a similar argument had weighed with
Mr. Ellis in interpreting 49, where Catul-
lus’s name is linked with that of another
great Roman, Cicero’s. He regards the lines
as eulogistic: but there is much to be said
for those who feel that the ring of the whole
poem is ironical. Just as in reading 56
we feel that Catullus was far too much
of a gentleman to exhibit himself exchang-
ing merely ribald jests with an unprincipled
scribbler, so here we believe that his sense
of humour would never have allowed him to
go down to posterity posing, awkwardly
enough, hat in hand, as Cicero’s very humble
servant. The poet and the orator must often
have met, and as far as we can judge their
relations can hardly have been very friendly.
Catullus was the friend of Hortensius,
Cicero’s rival: he was the lover of Clodia,
Cicero’s enemy: he was the assailant of
Vatinius, Cicero’s client. And though Mr.
Ellis (p. xxxiii.—iv.) has made it probable
that Catullus is not included in the ‘ canta-
tores Euphorionis’ of Z'wuse. iii. 45, still
Cicero’s well-known sneers at the νεώτεροι
(Att. 7, 2) and ‘novi poetae’ (Or. 161) sufii-
ciently indicate his attitude towards the
school of which Catullus was the exponent
and the pride. It may be fanciful to find
in the ‘Romuli nepotum’ of v. 1 an allusion
to the long struggle of the ‘ Romule Arpinas’
with Hortensius for the ‘regnum iudiciorum ’
(1 Verr. 35): but if Cicero’s name had been
publicly.coupled with Catullus’s in an eulogis-
tic poem such as this, it may be doubted
whether the orator would have written as he
THE CLASSICAL REVIEW. 29
has (Tusc. v. 63) : ‘adhue neminem cognovi
poetam ... qui sibi non optimus videretur.’
At best, there is a leer in the equivocal
‘tanto. ...quanto’ (v. 6), and equivoque has
no place in eulogy.
There are other instances, however, where
Mr. Ellis’s habitual ἐποχή seems to be carried
to excess. Of these the more important
have been long ago noticed by the reviewers
of his earlier edition, and only call for
remark now because they reappear stated
with additional emphasis, though hardly, we
think, supported by fresh proof. Mr. Ellis
still holds that the poet’s praenomen was
Quintus, and that he went to Bithynia in
65 B.c. On the former point, he has re-
examined the MS. evidence at some length
(p. xi.), and concludes that the ‘Q’ of D and
Pliny’s inferior MSS. may be traced to xii.
cent. and may represent a genuine tradition.
And these hypotheses outweigh with him
both the negative evidence of G. O. and
Pliny’s best MSS. which recognise no prae-
nomen, and the positive evidence of Apu-
leius and of Jerome, who has Gaius ‘ writ
large,’ so completely that he still reads
‘Quinte’ in 67, 12. This may pass.
The bloom of Catullus’s ‘florida aetas’
will smell as sweet under Quintus as any
other name. It is a different matter when
we come to the Bithynian journey. This is
the one definite episode in the life of Catullus
that has come down to us, it is the mse en
scéne of much of his most exquisite verse, it
had doubtless a direct bearing on the com-
position of the ‘ Attis’ and the ‘ Peleus and
Thetis ;’ and if we follow Schmidt in think-
ing that this appointment came as the
answer to the prayer of the divine seventy-
sixth poem, it gains fresh interest as dividing
the life and writings of Catullus into two
sharply contrasted periods. It is important,
therefore, to be right about its date. Mr.
Ellis’s view has been attacked, as is well
known, by Munro, though on grounds some
of which it is not easy to appreciate, (Crit.
and El. p. 46-7.) For even if the ‘ sudaria’
of 12 were sent in 65 B.c. it does not follow
that that poem was written in the same year,
and Munro’s appeal to 58 begs the ques-
tion of the date of that poem, which is inde-
terminate. But Mr. Ellis’s chronology raises
the question of the poet’s birth-year in a
manner which seems to demand a fuller con-
sideration than it has received. He admits
that Catullus must have been born either
in 87 or 84 B.c. Of these dates the latter
is surely the more probable. It alone can
be made to harmonize satisfactorily with the
statements of the ancient authorities, it has
the support of Munro and Prof. Sellar, and,
in addition to thei arguments, it may be
noted that it helps to explain Catullus’s life-
long companionship with Calvus, who was
born in 82. But if Catullus went to Bithy-
nia in 65 he cannot have been born in 84.
For the sixty-eighth poem, which precedes
the Bithynian journey, must then have been
written when the poet was at most eighteen,
and it is inconceivable that a lad of that
age, who was no prig either, could have
written ‘tempore quo primum vestis mihi
tradita pura est’ etc. (68, 15). Further, if
this poem belongs to 66 B.c. then the earlier
Lesbia-cycle must belong to the two years
previous: and we are to believe that a
country-bred boy of sixteen or seventeen won
the favour of the first lady in Rome and then
sang in lyric verse the praise of her ‘ whose
dead songster cannot die’—a supposition
which becomes only less incredible if the
birth-year is put back to 87. Lastly it should
be remembered that before Catullus went
to Bithynia he had his head-quarters in Rome
(68, 35), and a house of his own (‘ero’ 31, 12)
at Sirmio, while his father seems to have lived
at Verona (68, 27 : Suet. Jul. 93); and surely
these facts suit a man of seven or eight
and twenty better than a youth still in his
teens. We find nothing in Mr. Ellis’s book
we would so gladly see changed as his views
on that Bithynian journey, unless it be this.
Though he quotes, with apparent approval,
the appreciative comment of a true critic on
91,2: ‘in misero hoc nostro hoc perdito
amore fore,’ he still describes 73, 6 ‘quam
modo qui me unum atque unicum amicum
habuit’ as ‘probably the most prosaic
verse in Catullus.’ Far other to our feeling
is the effect of these ‘duriusculi,’ these
rasping syllables, which only Catullus can
propine ‘flammeos medullitus.’
We have scarcely scratched the surface of
the mine discovered by the noble labours of
Mr. Ellis. Our wish is merely to offer a
cordial and respectful welcome to a work of
the first moment in classical literature. At
the outset there seemed some danger of fall-
ing into terms of extravagant eulogy in
speaking of a book to which we owe so many
hours of pleasure and profit: as it is, we
have neared the opposite extreme. But a
critic very properly patronizes a book which
has taught him all the little he has gleaned
of the subject, and in any case Mr. Ellis,
after his thirty years’ ‘ fraternum sodalicium’
with Catullus, can afford to smile at the fret
of a modern Suffenus, ‘the irresponsible,
indolent reviewer.’
W. T. Lenprum.
90 THE CLASSICAL REVIEW.
THREE ESSAYS ON ASINIUS POLLIO AS A WRITER.
Untersuchungen zu Caesar u. seinen Fortset-
zern: insbesondere tiber Autorschaft u.
Komposition des Bellum Alexandrinum und
Africanum, von Dr. G. Lanperar, Erlan-
gen, 1888. 3 Mk.
Tue aim of this very ingenious monograph
is to prove that C. Asinius Pollio, the
famous orator, poet and historian, is the
author of the military diary which is
included among the works of Caesar under
the name of the Bellum Africanum ; that he
is also the rédacteur of the remains of Caesar
and his literary executor Hirtius, whose
death at the siege of Mutina (Apr. 27, B.c.
43) cut short his plans for the completion
and final publication of the Corpus Caesari-
anum.
It would be impossible in the brief space
at our disposal to give an adequate digest of
the argument: but its general course may
be indicated. It will be remembered that
in the received editions Books i.—vii. of the
De Bello Gallico are attributed to
Caesar, Book viii. to A. Hirtius, the De Bello
Civili (three books) to Caesar; and the
supplements to the latter, viz. the Bellum
Alexandrinum, B. Africanum, B. Hispaniense,
appear under the name of Caesar. But
even in antiquity the authorship of these
last three was a disputed point (Sueton. Vit.
Jul. ly.), some attributing them to Op-
pius, some to Hirtius. It was Nipperdey
who raised the question afresh in his
Quaestiones prefixed to his edition of Caesar
(1847), and sought to prove that B. G. viii.
and Bell. Alex. were by Hirtius, to whom
(and not to Oppius) the language of the
Eypistula ad Balbum, prefixed to the former
by its author, is appropriate. This epistle
plainly shows what Hirtius’s design was, viz.
to weave the Commentaries of Caesar into a
connected whole and carry them down to
the date of his death. Apart from the
actual state in which the literary legacy of
Caesar has come down to us, it is scarcely
conceivable that Hirtius could have accom-
plished his design in the short and busy
period which intervened between Caesar’s
death and his own. The facts of his con-
sulate and his illness must be borne in mind.
Nipperdey thinks that for Bell. Afric. and
Bell. Hispan. he had done no more than to
collect materials.
This hypothesis, though most valuable as
clearing the ground for further investiga-
tions, was not in itself fully satisfactory.
There are undeniable differences of style
between B. Alex. and B. G. viii. which mili-
tate against a unity of authorship. A
distinct advance was made when it was re-
cognized in part by Petersdorff, more fully
by Schiller, that the depositions of eye-
witnesses were incorporated in all parts of
the work and that variety of styles is due
to the variety of original sources. Schiller
conjectures that Hirtius worked over these
materials, without working them up into
organic unity: it is the connecting and
transitional chapters, he finds, which are
stamped with Hirtian peculiarities. It is in
reference to Bell. Alex. that Schiller formu-
lates his theory.
Here then Dr. Landgraf takes the ques-
tion up. He takes, one by one, all the
members of the Corpus Caesarianum: they
reveal to his fine perception of style the
touch of one, of two, and in one case (Bb. Alex.
ii—xxxiii.) even of three hands. He has
made the remarkable discovery that C.
Asinius Pollio was the eyewitness and
original authority on whom Hirtius relied
for those parts of the war in which he had
not himself been engaged; and that he, as
the next to Hirtius in Caesar’s intimacy,
undertook to complete, in outline at least,
the literary project which fate had not
allowed his friend to accomplish. The Sell.
Africanum he published exactly as he had
made it over to Hirtius. Its style, if the
text be rectified, is much better than critics
have allowed.
The evidence is partly external, partly
internal. (a) While no ancient writer con-
nects the name of Pollio with the authorship
of the supplements to Caesar’s work, he is
mentioned by Suetonius (/.c.) as a somewhat .
severecritic of the Commentaries: ‘parum dili-
genter parumque integra veritate compositos
putat.’ He did moreover write a History of
the Civil Wars himself (Hor. C. ii. 1). He
had served in the African war himself; had
also been governor of Further Spain from
the year B.c. 45, so that he may well have
furnished Hirtius with information upon the
Spanish events narrated in Bell. Alex. xlviii.
—lxiv. He was in a position to supply from
personal knowledge—conformably with his
severe veracity—a discreditable fact in B. @.
viii, 23, 3—6, 47, 48, 1—9, which Caesar
had suppressed in its right place, vil. 75. It
may be doubted if any other Caesarian had
similar qualifications for the task. (6) The
internal evidence’ of diction, &e. is the
᾿ THE CLASSICAL REVIEW. 31
strongest. In spite of the ravages of time,
enough has survived of Pollio’s composition
by which the theory may be tested: especially
three letters to Cicero, Ad Fam. x. 31, 32 and
33. We know from the Dial. de Oratoribus
and Quintilian that his style admitted
archaic and poetic elements, and even collo-
quialisms. Examples are (1) portendo, in the
Corp. Caes. only in B. Afr. 82, 2, and in Lp.
ad Fam. x. 32. 5. (2) In the B. Afr. the
following poetical expressions, swppetias ire,
tristimonia, sauciis decem factis: (3) nullum
vestigium ab aliquo discedere, Ad Fam.x. 31,
6., B. Afr. 73, 3. Dr. Landgraf certainly
makes out a strong case for Pollio as author
of the Bell. Africanum: besides style and
diction, the absence of partisan unfairness,
as in the treatment of Cato’s character, is
given in evidence. The very faults of the
work—its haste, triviality and want of
historical perspective—may be accounted for
if it was a soldier’s diary, not a history.
The main objection to the theory will be
this. Why is Pollio never mentioned in the
Commentaries, although he distinguished
himself so much in the field (Plut. Caes. 52) 2
For the same reason probably which led him
to grudge to the Caesarian remains that
effort which might have given them an
artistic uniformity and unity which they
lack. He had his own literary ambition, to
make himself the historian of the First Civil
War ; and he had no mind to forestall him-
self by a work, of which two-thirds of the
renown would belong to Caesar and Hirtius.
So he studiously dissociated his reputation
from their work, while he was too loyal to
neglect to make it worthy of theirs. But it
must be confessed that it is difficult to
understand how it was that his disguise
baffled all detection by critics, ancient or
modern, until Dr. Landgraf pierced it. We
understand now that Prof. Wéolfflin has in
preparation an edition of the Bellum Afri-
canum, to be published, missis ambagibus,
under the name of C. Asinius Pollio. Dr.
Landgraf promises us (p. 78) a future dis-
cussion of the contribution made by Pollio
to the Bellum Civile: pointing out for the
present the Pollionisms in &. C. iii. 108-112,
the opening of the Alexandrine war. No
student of the diction of Caesar can afford to
neglect the contributions to the subject
which enrich Dr. Landgraf’s pages, e.g. the
valuable note on postquam with subj., p. 20.
But it is not true that the historic infin.
is used by Cicero ‘only in the Verrines and
the Letters’ (p. 41): v. Cluent. §$ 59, 177 ;
Pis. § 69; Phil. xii. ὃ 1; Acad. pr. ii. §§
1G;
Ueber den Sprachgebrauch des Asinius Pollio
in den bei Cic.ad Fam. x. 31—33 erhaltenen
Briefen mit Berticksichtigung der bet Quin-
tilian, Seneca, etc. tiberlieferten Fragmenta,
von J. H. Scumatz,
Wolfilin, Archiv, vi. pp. 85—107, ‘ Ueber
die Latinitit des Asinius Polio.’ Miinchen,
Beck, 1889. Mk. 1.40.
A review of Dr. Landgraf’s treatise would
be incomplete without a short notice of these
two essays, the former of which discusses
the diction of Asinius Pollio as it appears
in the Letters, the latter that of the
Bell. Africanum. In their general con-
clusions they are at one: that Pollio (or
Polio as Wolfflin prefers to call him) did not
give allegiance to the Latinity of Caesar and
Cicero, but consciously copied other models,
and stands nearer to the speech of the ar-
chaising Varro, and the abrupt and some-
times colloquial Sallust. Surely it is curious
that he himself criticised the archaisms of
the latter—‘Asinius Pollio in libro quo
Sallustii scripta reprehendit ut nimia pris-
corum verborum affectatione oblita’ (Suet.
gramm. 10), while he allowed himself the use
of such archaic words as the following
pollicitatio, commeatus ‘a way,’ superesse
‘survive,’ guiritare (which has lived on in
popular speech to become Fr. crier, Ital.
gridare), palpari, all cited by Schmalz :
suppetiae, grandis ‘large,’ rapsare, tristt-
monia,; satagere infatuo conpotare (these three
perhaps old-fashioned colloquialisms), cited
by Wolfflin. It seems possible that in his
maturer work Pollio was less unwilling to
part with some of the angularities of his
style, though still the declared foe of Cicer-
onianism gone mad (Quint. x. 2. 17, si quid
modo longius circumduxerunt, iurant ita
Ciceronem locuturum fuisse). The task
which at present occupies Landgraf and the
others is to prove that Pollio forms a link,
in the development of Latin prose, between
Varro and the historians Livy and Tacitus.
They may afterwards hope to consider the
growth of Pollio’s own style in the frag-
ments preserved to us.
Both writers show exhaustively that there
is alike an archaic and a poetic element in
the style which we are considering: our
author uses communibam, a form which
‘Cicero would have barely touched in kid
gloves’ (Wolffl. p. 88), an zsolated form (if
we can trust our MSS.) which Pollio may
have picked up from the sermo castrensis ;
he uses vectigaliorum, and a few other
anomalous forms. It is further shown that
he deals in colloquialisms, such as the use of
32 THE CLASSICAL REVIEW.
compound words in the sense of the simple
forms (commendare se fugae, ex promerito,
inibi, depugno =pugno). Lastly, in his earlier
work at any rate Pollio cannot be acquitted
of the charge of occasional slovenly writing.
The abl. absolute is carelessly used: B. Afr.
24, 2, Pompeio adveniente oppidani usque eo
passi propius accedere, donec cet. ; and even
Ad Fam. x. 31. 4, nam quibus commeatibus
invito illo per illius provinciam legiones
ducerem? Κ΄. also B. Afr. 63.5. But for the
sake of distinctness, such a thing is not un-
known even in Cicero: v. Roby, 1247. Again
we find a sort of Homeric construction in
relative clauses, B. Afr. 97. 3, Leptitanos
quorum bona Juba diripuerat et ad senatum
questi...sua receperant. Other idioms, such
as dubito with acc. and inf., opto with inf.,
studeo with ace. and inf. (Schmalz, 26-—28)
seem to be due to the influence of the older
poets. Butis not the pregnant use of in
(Ad Fam. x. 32. 4, in victoria = 51 vicisset)
commoner in Cicero than Schmalz implies on
p. 201 Γ΄ my note on Cluent. ὃ 32.
An examination of the facts adduced in
these two treatises will lead, I think, to the
conclusion that even in his Letters Cicero was
much more of a purist, than Pollio in his
most studied efforts.
W. Y. Fausser.
NETTLESHIP’S CONTRIBUTIONS TO LATIN LEXICOGRAPHY.
Contributions to Latin Lexicography. By
Henry Nettiesuip, M.A., Corpus Profes-
sor of Latin in the University of Oxford.
Oxford. Clarendon Press. 1889. 215.
Tuts book, like the larger work of Pro-
fessor Key which appeared a year ago,
contains the portion of a Latin lexicon
which was more or less advanced when
death in the former case, and disinclination
or other engagements in Professor Nettle-
ship’s case, interfered to prevent the comple-
tion of the undertaking. Necessarily they
appeal not so much to the general public,
even of classical scholars, as to the makers
of lexicons or to special students, but both
were worth publishing, and will doubtless
contribute to improve the Latin dictionaries
of the future. Both lay great stress on the
early Roman literature, both are the work
of independent scholars, and both testify to
a great deal of thought and labour. Profes-
sor Key no doubt aimed specially at setting
forth his peculiar views of etymology. Pro-
fessor Nettleship has a bit of a bias for
another side of the investigation. He has
paid great attention to the Latin glossaries,
and among numerous other recondite ad-
ditions has quoted many of them in the
articles which form the present work. Such
sources have some sort of value for lexicons
on a large scale which profess to give the
materials, important or not important, for
the history and signification of a Latin word.
I have myself a poor opinion of Latin glos-
saries: they are evidence of very uncertain
character—uncertain in their age, in their
author, in their ultimate origin; they may
be witnesses to a regular use and meaning,
they may record merely the momentary
fancy or whim of a speaker or writer, or
they may embody a misunderstanding of the
reporter or a blunder of a copyist. And
even the better authenticated productions of
Latin grammarians are so uncritical, that I
think them often much overrated testimony.
Unless we have the context and the circum-
stances, a reported use of a word requires
special corroboration from general analogy
or probabilities before it is entitled to any
considerable weight. The desperate eager-
ness of scholars to throw some fresh light
on obscure or disputed matters seems to me
often to result in a very factitious import-
ance being attributed to the dustheaps of
antiquity.
About two-thirds of the present volume
is occupied with the letter A ; the rest con-
tains articles on various words, which for
some reason or other attracted Professor
Nettleship’s attention. He tells us he has
included nothing in this volume which, in
his own judgment, is not a necessary im-
provement of Lewis and Short’s Dictionary,
either by rearrangement of the matter, or
by novelty in explanation, or in instances,
or by correction of references, or by the
supply of omitted words. Possibly he has
in this adopted the right principle, but the
result is that one never quite knows whether
any article should be regarded merely as a
supplement and correction of Lewis and
Short, or really representative of what Pro-
fessor Nettleship considers a proper and
adequate treatment of the word. However
the articles are always scholarly and clear,
THE CLASSICAL REVIEW. 33
and many exhibit a very rational and suit-
able arrangement of the matter. I have no
intention of detracting from a general ap-
preciation of the book, if I notice a few
cases where I think Professor Nettleship has
slipped. Most of these small points relate
to legal expressions.
On p. 19 he says the title de accessionibus
of Dig. Book XXII. includes uswrae and
Jructus. On the contrary the title contains
all three words, and accesstones are neither
usurae nor fructus. Below he speaks of
accessio temporis as if it were more technical
than it really is, and defines it as ‘the
period of time reckoned in order to make up
the year required to get the benefit of the
interdictum utrubi, Gai. TV. 151.’ But a year
was not required for this purpose : the ques-
tion was simply which of the litigants had
been longer in possession within the period
of a year immediately preceding the suit.
On p. 252 he asks what is the meaning of
adprehendit in Dig. XL. 7,1. 2 pr. qui statu
liberi causam adprehendit. The meaning is
simply ‘he who gets the position of a statu
liber (slave contingently freed).’ On p. 203
he translates annalis ‘annually recurring,’
but in two of the instances quoted the word
refers to the period, not the recurrence, of
a year. So Varr. 1. 1. 1, 27, where Varro
speaks of the annale tempus quod sol circuitu
suo fintt. In the other, Dig. XV. 2, 1. 1,
an action is said to be temporaria id est
annalis, ὁ.6. 1 must be brought within one
year from the right of action accruing. On
p. 263 Professor Nettleship misunderstands
in some degree the nature of book-obliga-
tions (htterarum obligatio), and thus after
rightly explaining arcaria nomina (Gai. III.
128) as ‘debts contracted by the actual
payment of ready money to the debtor,’
hesitatingly adds, ‘without any evidence
given in writing.’ But the presence or
absence of evidence in writing does not con-
stitute the difference between these classes
of debts. <Arcaria nomina were doubtless
often accompanied by documentary proof,
although their obligation rested on the actual
cash payment, not on a bond or on a
book entry. In the other case the entry
in the books of the creditor, or of both
creditor and debtor, was the very ground of
the obligation ; and for this reason, because
the loan or sale or other transaction, which
may have preceded, was by the agreement
of the parties merged in the book entry, the
former debt, if any, was extinguished, and
the only debt now existing was that created
by the book entry, and was attended only
by the incidents belonging to that. The
ΝΟΣ XXX. VOL, IV.
nearest English analogy is where previous
transactions involving debts and credits are
summed up and superseded by a bill of ex-
change for the balance, drawn by one and
accepted by the other party.
On p. 440. dispunctio is said to mean
‘rewarding according to deserts,’ and dis-
punctor, ‘one who allots payment.’ This
is, 1 think, a mistake. Dispungere is ‘to
check’ or ‘examine accounts’ and is so
used in the Digest: e.g. dispungere est con-
Jerre accepta et data, Dig. L. 16, 1.56: ef. XL. 7,
1. 6, 8 7; XLII. 5, 1. 15, 81; and the same
meaning appears to me clearly applicable to
the passages quoted from Tertullian (who,
be it remembered, was probably a lawyer
himself), Apol. 18, suscitatts omnibus ab
initio defunctis et reformatis et recensitis ad
utriusque meriti dispunctionem, * All the dead
from the very beginning being raised and
reformed and reviewed for an account to be
taken of their good or evil deserts,’ and so
in Mare. V.12; IV.17. Dispunctor in the
inscriptions (in some we have Disp. only) is
I presume simply a public accountant.
On pp. 68 and 456 we have the old word
Jatis, found in adfatim, explained as ‘ weari-
ness.’ I long ago suggested that it meant
‘yawning,’ cf. fatiscere, fatigare. 1 doubt
also whether my account of noxa and ob-
noxius (see Introduction to Justinian’s
Digest, pp. 132—134) has fallen under Prof.
Nettleship’s attention. He seems to me to
invert the true order, when he gives noxa the
meanings (1) punishment (2) offence (3) harm
instead of (1) harm (2) guilt (3) amends. I
cannot see why he is not contented with the
old and perfectly easy derivation from nocere,
nor why he suggests that obnoxius is not
from nocere and ποῦ, but from a verb ob-
mectere which he has found in Festus and a
glossary. I take obnowius to be formed from
ob noxam and to have passed naturally from
the specific meaning of ‘ liable for a hurt’ to
the general meaning of ‘liable.’
Attribuere in ofticial or technical language
is used of giving an order on a person to pay
money, the person being put in the accusa-
tive and the money (sometimes) added in the
genitive. This usage Prof. Nettleship has
noticed. But he has misunderstood the
analogous use of pecunia attributa, ‘money for
which an order is given,’ e.g. Cic. Verr. 1. ὃ
13 pecunia attributa numerata est, where he
strangely takes attributa as ‘paid over.’
What then is nwmerata? The most illus-
trative passage is Verr. 111. ὃ 165, cum posita
esset pecuni apud eas societates unde erata
attributa, binis centesimis faeneratus est,
deinde permultis civitatibus pro frumento
D
94
nihil solvit omnino, ‘when the money was
lying with the companies on whom an
order was given him for its payment, he
lent it out at two per cent. and paid to town
ABBOTT'S LATIN GATE AND
The Latin Gate: a First Latin Translation
Book. By Epwin A. ΑΒΒΟΤΊ, D.D., Head
Master of the City of London School.
London. Seeley & Co. 1889. 3s. 6d.
In this book, which we may call Dr.
Abbott’s legacy to classical education on
retiring from the active practice of his pro-
fession, we have what is really an educational
experiment of some interest and novelty.
The author recognises most truly, as the
leading defect of the teaching of Latin in
middle class schools, that the short time
available for the study is most wastefully
spent on Composition and Grammar, with
the result that at the end the pupils cannot
translate easy Latin at sight. The cure for
this evil he would seek in more rapid and
economical methods of teaching to read the
language : less grammar learning and prose
writing, which, useful enough for those who
mean to pursue a larger scheme of study,
lead practically to nothing for those who
leave off at sixteen : and the insistence, in all
public examinations, on translation at sight
as the most searching and useful test. Of
these lines of improvement the last depends
on the examining bodies; the second is
obviously in the control of the schools. This
book is intended by Dr. Abbott as a con-
tribution towards the achievement of the
first, namely the improvement of method in
teaching boys to read Latin.
The first part of the book is practically
an attempt to help the mastery of the
vocabulary, by keeping the beginner’s
attention on the identity of Latin words
with the corresponding French and English.
All teachers no doubt do this at times, and
in a desultory way: but Dr. Abbott well
suggests, in his second and third chapters,
- how it may be done more systematically. If
these chapters are used, as the author plainly
means, not as lessons to be learnt (which
would be cram of a dreary kind) but by the
teacher as a guide to method, and by the
pupil only for reference, we think the result
would be excellent. The Latin vocabulary
would be mastered more easily and securely :
THE CLASSICAL REVIEW.
upon town nothing whatever for the corn
they supplied.’ (This use of attribuere is
noticed in Mommsen’s Staatsrecht, ii2 444.
n. 2.)
ἘΠῚ Jie ROBY,
POSTGATE’S SERMO LATINOS.
and there would be something like a real
understanding gained of the Latin element
in French and English.
The next part is called ‘ How to translate
a Latin sentence’: and is virtually an
attempt to give help in the first real problem
of the reader of Latin, the solving of the
structure. Dr. Abbott begins with the
usual rule, Find the verb, and then proceeds
to deal practically with the chief common
difficulties in relation to the verb. One
suggestion here is valuable : that the student
should drop this rule as soon as he can, and
‘begin to translate mentally straight on,
making allowance for the Latin way of
putting things.’ A propos of this, there is a
good deal about order: perhaps (as Com-
position is out of the question) rather too
much ; but we are glad to see a reference to
Professor Hale’s brilliant essay on the Art
of Reading Latin, from which Dr. Abbott
quotes an amusing illustration of the import-
ance of orderin a long sentence. Professor
Hale takes a passage from Livy, and shows in
detail how the apparently arbitrary disregard
of the logical order is really due to the fact
that the order is everywhere natural. When
this is perceived the student gets great
encouragement to try translating ‘mentally
straight on’: and nothing helps so much to
rapid reading, z.e. to reading Latin as a
language.
The rest of the book is reading lessons
graduated in length with full notes. First
proverbs and short sentences: then longer
sentences: then a book of Phaedrus, and lastly
a book of Caesar. The main novelty here is
that the sentences are carefully chosen to be
either striking sayings or anecdotes. For
ourselves we much prefer the anecdote for
the young beginner. Many of the striking
sayings simply bore the boy, who has no ex-
perience of life to make him appreciate them :
others disgust him by their medicinal flavour :
and many lose much of their effect and even
their meaning without the context. What
can a boy care for ‘vox populi vox dei,’ ‘ars
longa, vita brevis,’ ‘silent leges inter arma,’ and
the rest of them? But he takes some
THE CLASSICAL REVIEW. 90
interest in Martial’s statement (for example)
that ‘he would rather have a tank than a
vineyard at Ravenna’ (where water is bad
and scarce) : and complains that an innkeeper
‘had cheated him by selling him neat wine,
when he had ordered wine and water. Or
again he will relish Catulus’ reply to the bad
speaker who asked him whether he had not
moved the feelings of the audience in his
peroration: ‘Certainly,’ said the cruel
Catulus: ‘nobody I should think was so
hard-hearted as not to feel great pity—
for your speech.’ In Dr. Abbott’s second
and third series there are many excellent
anecdotes like the above, which tide the
learner over the initial flats and bars of Latin
reading.
We may note a few slips—though, as the
first edition is only issued to teachers, for
comment and correction, they have perhaps
already been emended.
The note on accent (ὃ 64) is confusing :
syllables were surely retained because they
were accented, not because the accented syl-
lable was significant.
The phrase ‘national instinct’ in § 67 is
not happy: the syllables were simply modi-
fied by wear and tear in spoken use.
§ 70, plaisir for plaire.
§ 76, on (in the phrase on dit): the learner
would think on confined to that use.
§ 91, peignant from pingentem is not really
a ease of transposition: the y-sound which
has supplanted the g is still after the n.
§ 103, the form mien should be mentioned
as well as mon.
§ 131, iv. fingere misprint for pingere.
§ 185, ‘The subject is sometimes the action
implied in the verb’ is not a clear account of
sie ttur.
But these are trifling blemishes in what is
an interesting, and likely, if properly used,
to be a valuable schoolbook.
Sermo Latinus ; a Short Guide to Latin Prose
Composition. By J. P. Posteate. Pp. 90.
Macmillan. 2s. 6d.
Tus unpretending little book, consisting
as it does of some sixty pages of advice and
discussion, followed by forty-one exercises,
we commend most heartily both to those
who have to teach, and those who wish to
learn, Latin Prose Composition. Without
professing to be exhaustive, Dr. Postgate
has succeeded in being, in the best sense,
practical: and we feel sure that both
teachers and learners will find his remarks
helpful and suggestive.
The first chapter is a brief statement of
the educational value of Latin Composition.
We agree with the author’s main point, the
value in training of that recasting of a
thought which translation into an ancient
language necessitates. Perhaps he is claim-
ing too much when he says that ‘ controver-
sies would often dissolve into emptiness’ if
disputants would put them into Latin
Prose. It would often do as well to translate
them into plain English : and even Romans,
in spite of their ‘ direct and concrete expres-
sion, were not exempt from the common
frailty of quarrelling about nothing.
The two remaining chapters, however,
constitute the valuable part, and the main
part, of the book. The first one (chap. ii.)
is addressed to teachers. The author touches
with great felicity the first great difficulty
of the teacher, when he feels that a version
offered ‘won’t do,’ but is unable to say
why. Dr. Postgate justly says he must
neither disregard this feeling nor think it
infallible. It is perhaps a pity that we are
not more decisively warned that the teacher
should never be content till he has analysed
and made definite this instinctive criticism.
Thesuggestions on retranslation are good and
stimulating. But the most striking part of
this chapter is the passage on class-teaching of
Latin Prose: and Dr. Postgate’s hints will,
it may be hoped, really help many a reader
to succeed in this most difficult branch of
the art, in which the writer is evidently
a master. Particularly would we refer
teachers to the admirable example, too long
to quote, where he discusses and translates a
difficult passage from Froude.
The third chapter is addressed mainly to
the pupil, but will be found very useful for
teachers too. With the remarks on the use
of dictionaries we heartily agree. In deal-
ing with the difference of expression in the
two languages the main point is well brought
out and excellently illustrated, that the
Latin always clings close to actual concrete
fact. Indeed this principle forms so large
a part of what the student has to learn, that
we would gladly have had more detail on
this head, even at the cost of some of the
remarks about order. It is to this prin-
ciple that is really due the objection of
Latin to what we may call the ‘ picturesque
alias’ (like ‘ the conqueror,’ ‘the iron Duke,’
‘England’s greatest general’) which Dr.
Postgate ascribes to the poverty of Latin.
Latin is quite able, of course, to render all
these aliases of ‘the Duke of Wellington’ ;
but, as they are mere flourishes, its practical
instinct refuses to have them.
D2
90
The expansion of ‘ Balbus is building a
wall’ is amusingly and cleverly done, and
will be found instructive: but a playful
protest may be permitted against the por-
tentous sentence into which it finally grows,
lest the learner should be beguiled into
taking it as a model.
In treating of order, we should have liked
to see more about the variety due simply to
the desire of pleasant sound, and the wish to
avoid dulness. Perhaps also more thorough
analysis and illustration might have been
given of the common English tendency to
what we may call ‘concealed oratio
obliqua’: the tendency, that is, to present
what is really the view or the suggestion of
one of the actors in the narrative as part of
the narrative itself. For example:
‘The general did his best to restrain the
ardour of his troops. Over-haste was likely
to be as fatal as delay. The plot was
hardly ripe: and to venture at once would
easily end in losing all. Accordingly he took
measures, ete.’
The second and third sentences here are
really of course the general’s, and in Latin
must be oblique. The instance from Gibbon
which the author gives on p. 36 (an admir-
able piece and well-discussed) is less likely
PLESSIS ON GREEK
Traité de Métrique grecque et latine, par
F. Puessts. Paris, Klincksieck. pp.
X00, eo, Li,
Tuts little work forms one of the useful
series in which Riemann’s Latin Syntax
and some others noticed in this Review have
already appeared. It is based, in a great
measure, on the Métrique grecque et latine of
Havet and Duvau (1888), the Métrique of
Gleditsch in Iwan Miiller’s Handbuch, and
the well-known works of Lucien Miiller and
Wilhelm Christ. But the matter is well
digested, and the subject is set before the
reader with clearness and precision. After
some preliminary ‘notions générales,’ the
author proceeds to treat of the construction
of the hexameter and pentameter in both
languages, and then of the various iambic and
lyric metres. In a series of short excursuses
he deals with the position of words in Latin
verse, the relative proportion of dactyls and
spondees in the first halves of hexameters
and pentameters, the prosody of archaic
THE CLASSICAL REVIEW.
to mislead the pupil than the passage we
have written above, which is one of a common
kind in modern English.
Dr. Postgate’s little book is certain to
require soon a new edition. We should like
to suggest that he should enlarge its scope,
so as to give us more detailed hints from his
own varied and carefully recorded experi-
ence on many points where he is now con-
tent to quote from other well-known works
on composition. Might we also suggest,
that in this case he should suppress the
rather scornful note on p. 6, against the
proposed substitution of French and German
for Classics in the case of certain students ἵ
The grounds for this proposal lie far outside
a mere comparison of the languages: and
they commend themselves, as Dr. Postgate
must be aware, to many who are earnest
believers in the Classical tongues as educa-
tional instruments. Even if it were not 80,
the modernisers would not be converted by
the bald assertion that ‘it will not be long
before Englishman will (except for special
reasons) as little think of learning either
[French or German] as they do now of
learning Dutch or Welsh.’
AL:
AND LATIN METRE.
Latin including the Saturnian metre, and
the law, commonly known as Meineke’s, of
the division of Horace’s odes into quatrains.
The correctness of such a division he im-
pugns. ‘There is lastly an Index Rerum so
constructed as to include a sufficient explana-
tion of the grammatical and metrical terms
employed.
The scientific method which characterizes
good modern handbooks, even when the
subject is versification, has something in it
rather chilling to the young student. It
is formidable to see tabulated, as here (p. 80),
the exact percentage of elisions in Catullus,
Propertius, and the rest ; or to have caleu-
lated decimally the permutations of dactyls
and spondees admitted by those poets in the
open feet (pp. 281 8φῳ.). With some minds
the result might be a distaste for efforts at
verse composition, like the disinclination
felt for reading Shakespeare after a study of
end-stopt lines and weak endings. But the
process is inevitable, and in the manual
before us the thing is certainly done well.
THE CLASSICAL REVIEW. 37
One advantage of treating Greek and
Latin metres side by side is that the con-
tinuity of development pervading them can
be better seen. The way the pentameter
forms a connecting link between the κατὰ
στίχον and the lyric metres (p. 100), the
gradual restrictions on its last syllables
(p. 116), and the slow but subtle influence of
accent (pp. 295-6), are clearly brought out.
The subject of accent, indeed, as affecting
and affected by metre, is an interesting one,
and one which has not received in English
text-books the attention it deserves. Dis-
carding the use of the terms arsis and thesis,
as being now generally understood in a sense
exactly contrary to their original one, the
author adopts a notation (p. 3) slightly dif-
fering from that of Havet and others to
designate the temps fort, or stress, as distin-
guished alike from quantity and the tonic,
or true, accent. The first of these three
he makes equivalent to the modern
accent (p. 34), while the last was
simply a momentary raising of the pitch, a
‘note musicale plus élevée.’ Such lines as:
Italiam fato profugus Lavinaque vénit
show how unaffected the metrical stress was
by the accent proper. The fact no doubt
remains that in the last two feet of the
hexameter we often find the stress or temps
fort and the accent coincide, as in ‘rura
manebunt,’ ‘lumine caelum,’ and such like
common types. But Havet (Métrique, p. 60)
will not allow this to have been intentional,
and he decides that ‘le réle de |’ accent est
absolument nul dans la versification latine.’
Weil and Benloew, in their Théorie générale
de 1 Accentuation latine (a treatise unfortu-
nately now out of print), come to much the
same conclusion (p. 241), on the ground of
the tendency of the Latin accent to coincide
with the stress (the Romans having for
example no words accented like Αἰσχύλος),
and of stress again to coincide with quantity.
In the long run accent carried the day, and
made possible the metres of the Christian
poets.
Not the least valuable part of the book is
that in which the writer treats of archaic
prosody (Hacurs. iii.), especially as affecting
the scansion of Plautus. The book is care-
fully printed, and we have noticed very
little to add to the list of errata. On p.
222, in the first line of note 2, ‘sur la
finale’ should be ‘sur Jinitiale,’ and the
last number on p. 279 should be 42 instead
of 41. We think the work will be found
very serviceable in the higher forms of
classical schools.
J. H. Lupton.
VIERECK’S OFFICIAL GREEK.
Sermo Graecus quo Senatus populusque
Romanus magistratusque populi Romani
usque ad Tiberi Caesaris aetatem in scriptis
publicis ust sunt examinatur : scripsit
PauLus VIERECK, Dr. Ph. Gottingae,
mMpcccLxxxviil. (5 Mark).
THE subject of this careful and scholarly
prize-essay (‘ praemio regio ornata’ ) may at
first sight appear to be of narrow range, but
the reader will find it opening out in various
interesting directions. When Rome became
mistress of the Eastern world, she had to
treat with her subjects in their own tongue,
and that tongue was Greek: Graecia capta
Serum victorem cepit. In the West, indeed,
the Roman official might usually employ his
native Latin, although probably Greek was
the channel of communication between
Rome and Carthage in the times of Hannibal ;
while in Magna Graecia we find Senatus-
consulta inscribed in Greek, and not in Latin,
late in the first century a.p. (C.l.G. 5836,
5838, 5843). But in Greece proper, and
over those large tracts of the world which
had been Hellenised under the Diadochi,
Greek was the official language: even in
Palestine the Roman decree or treaty was
inscribed in Greek as well as Latin (p. xii).
The writer first addresses himself to the
task of collecting and arranging the texts of
all the inscribed documents which come
within his scope. These are set out in full,
the readings being subjected to careful
criticism. First come ten Letters from
Roman magistrates, the earliest being
Flamininus’s letter to Cyretiae (circ. 196 B.c.)
and the latest (p. 47) a letter of Cn, Lentulus
Augur to Nysa (s.c. 1). Next follow eleven
Senatusconsulta, most of them in excellent
preservation and of considerable length, their
number being swelled by several more in
the Addenda. A third class is formed by
the Treaties. None of these documents are
88 THE CLASSICAL REVIEW.
new, and many of them have long been
familiar. But they gain greatly in interest
by being thus grouped together, and the
student of Roman history may perhaps find
in this collection some documents that have
escaped his notice amid the scattered liter-
ature of epigraphy. Some of the Senatus-
consulta were first published by me in Part
111. of the Greek Inscriptions in the British
Museum, and if this Essay had then lain
before me it would have saved me much
trouble in restoring the texts. The collec-
tion might have been greatly enlarged, had
the author wished to proceed further than
the reign of Tiberius. Imperial letters
addressed to the Greek cities are being
brought to light in almost every excavation.
Some are the merest ‘ notelets’: but Greek
servility inscribed them all. At this day
however they have a certain value, if only
in the chronological data furnished by their
headings. Few however are so curious as
the document discovered at Acraephiae last
year (Bulletin de Correspondance Hellénique
xii, 1888, p. 510), which gives us a verbatim
report of the high-flown address of the
Emperor Nero to the Greeks at Corinth, in
proclaiming the liberty of Hellas, Nov. 28,
4.0. 66.
Our author, after placing the documents
before us, proceeds next (pp. 55 foll.) to
criticise their language, under the various
heads of orthography, inflexions, syntax,
vocabulary and style. There is no doubt
whatever that these Romano-Greek docu-
ments received their Greek dress at Rome,
and were transmitted in Greek to the Pro-
vinces. In grammar and diction they differ
but little from each other, or from the
Greek commonly spoken by educated persons
in that day. Latinisms, indeed, of idiom
abound, but the number of Latin words
bodily transferred into the Greek tongue is
remarkably small: nearly every Roman
official term received at once its official
Greek translation. We must remember
that it was as essential for a Roman states-
man, even under the Republic, to speak
Greek as for an English public man to-day
to speak French (ep. Livy, xlv, 29); it was
usual for Greek envoys to address the Senate
in their own tongue. The extant letters of
the Roman politicians afford an amusing
evidence of the extent of their acquaintance
with Greek. Flamininus’s Greek style is
crude, involved and obscure ; his diction is
commonplace, like that of Polybius (p. 75),
and not free from Latinism and _ bad
grammar. Mark Antony, whom Cicero so
unmercifully banters about his rhetorical
studies, affects a verbose and grandiloquent
manner, not always strictly grammatical
(p. 77). ‘The literary revival of the
Augustan age is reflected in the elegant
Greek of Augustus’s own letters (p. 78): and
very curious ones they are, as illustrating
Provincial life and government.
In the course of this discussion of points
of style, some interesting illustrations occur
of the change of meaning which some Greek
words underwent in later days: e.g. ἀνωτέρω
‘before’ of time, or ‘above’ in a book;
καταχωρίζειν ‘to put on record in a book’ ;
φιλάνθρωπος of kindness generally ; καταλογή,
‘distinction,’ ‘notice,’ ‘honour’; etc. These
pages (72 foll.) are of value to the student of
Hellenistic Greek. I doubt the statement
made on p. 56 about the revived observance
of the iota adscriptum in the Augustan age
giving rise to its insertion where it was not
wanted. Long before the Augustan age, in
purely Greek documents of the time of the
Diadochi, such otiose insertion of the iota
was not;unknown: Mr. Roberts agrees with
me in reading ὁρκῷσ[αι] in line 20 of the
Lygdamis inscription. Viereck does well,
however, in declining to call forms like
ἀπήλθοσαν, εἴπασαν ‘ Alexandrian,’ since they
are found equally in other parts of Greece
(p. 59).
A separate discussion is reserved (p. 85)
for the Greek version of the Marmor
Ancyranum, which departs so far from
ordinary standards of grammar and style,
that Nissen believed it to have been made
from the Latin by a Gaul of Ancyra.
Viereck suggests that the inferior style of
this document may readily be accounted for
by its length and the great complexity of
its contents. These afforded more oppor-
tunity for the exhibition of ignorance than
the much more simple and formal documents
we have hitherto been discussing. He con-
cludes that the Greek of the Marmor
Ancyranum was drafted by a Roman official
in Rome itself, and was transmitted to
Galatia, by order of Tiberius upon the
request of the people of the province, to be
inscribed on the walls of temples dedicated
to Caesar-worship.
The examination of so many authentic
inscribed documents qualifies the author to
proceed yet further, and to discuss the
specimens of Senatusconsulta and official
letters which occur in Polybius, in Josephus
and in the Book of Maccabees. Internal
evidence satisfies him that Polybius had
before him Greek originals of public docu-
ments (p. 90). Josephus is convicted of
gross carelessness, and something worse, in
THE CLASSICAL REVIEW. 39
his handling of Senatusconsulta and official
letters borrowed by him from the collections
of others. But there is no ground for doubting
the substantial accuracy of the documents
themselves, which were mostly borrowed
from Nicolaus Damascenus, though little
trust is to be placed in their headings and
the titles assigned to the Roman officials in
them ; these are due to the unscrupulous
pen of Josephus.
The essay is made the more useful by
some good indices. I have noted only one
glaring misprint; p. 7, line 11: templum
Veneris Ephesiae, for Dianae.
E. L. Hicks.
PLOIX’S LA NATURE DES DIEUX.
La Nature des Dieux. Par CHARLES PLOIX.
1888. 10 fr.
OnE opens this book, which consists of only
470 pages, with a pleasurable expectation,
first because it is written by a Frenchman,
and therefore it is not likely to be dull or
obscure, and secondly because of the title.
The student of the classical religions might
hope that he will here find—not one more
theory added to the endless number concern-
ing the origins of those religions—but a work
which we in England especially need on the
actual religious ideas of Greece and Rome in
the historic periods. But the whole book is
a theory of origins, and its title is an illu-
sion; it is also an astonishing, perhaps an
expiring, effort of the familiar school, in
which we have all been trained, of philolo-
gico-physical mythology. In his preface M.
Ploix disclaims style and aspires merely to
science. But the book has of course the
usual French excellences of style, piquancy
of expression and lucidity, while as regards
science it reminds us of Mr. Lang’s treatise
on ‘the Gladstone Myth’: only M. Ploix
means his book to be taken very seriously.
It is in fact nothing less than a key to all
the polytheistic religions ; for what he dis-
covers of the Greeks and the Romans he
maintains must be true of the whole human
race. Thus at the outset we are startled
with the enormous assumption—with the
proof of which he does not trouble himself—
that ‘la marche de l’intelligence humaine a
du étre partout la méme.’
Plurality of causes is excluded at a single
stroke. Ancestor-worship, the sense of the
Infinite, personification of moral ideas, and
other possible explanations, cannot be applied
to any religion of any branch of the human
race if they are found not to apply to the
religion of Greece and Rome. If one could
believe this, one would at least save oneself
trouble. With this advantage to start with
M. Ploix works out the great single idea
with which his book is inspired, namely, that
all the Greek and Latin divinities except
Zeus are personifications of the twilight.
He has at least maintained his originality,
for the twilight is the one department of
nature that has not been over-run by his
fellow-mythologists. We have become very
familiar with the apparitions of the sun, the
moon, the wind and the lightning, etc., in
the forms of Apollo, Dionysos, Hera, Athene,
and many others, but Apollo, Demeter, Pro-
serpine, Eros and Ares as dawn-divinities of
twilight are novel characters. M. Ploix is
very contemptuous of solar theories, and
holds some very peculiar views about the
sun. He thinks that primitive man paid
very little attention to it, and did not regard
it as the cause of light, and supposed it was
a very slow thing, and that therefore he
could not have personified it as a fast-run-
ning hero such as Apollo or Hermes. ‘Le
soleil ne parait avoir été anthropomorphisé
chez les Grecs et les Latins.’ This is a
strange statement in the face of the worship
of Helios at Rhodes and elsewhere, and his
very human activity in many Greek legends.
However, though he disregards all solar,
lunar or astral explanations, he maintains
as an indisputable view that the divinities
were personifications of celestial phenomena:
in the first place, because Devas and Divi
and Zeus contain a root meaning bright:
therefore they must be sky-divinities or
deities of light. We are familiar with this
argument through Professor Max Miiller’s
writings ; but modern philologists by no
means accept this conclusion. For the same
root may be used for the expression of two
independent though cognate ideas: of two
derivatives from the same root the one may
denote a ‘ bright being,’ the other the ‘bright
sky,’ and yet the bright being need not be a
personification of the sky. And after all the
Greeks did not speak of Divi but θεοί, and
40᾽ THE CLASSICAL REVIEW.
what else this word means besides ‘ gods’ no
one yet knows, though M. Ploix thinks he
does.
A second argument that M. Ploix urges
is the way in which he supposes polytheism
arose from fetichism—for he admits promis-
cuous fetichism as a prior stage. There is
some truth and ingenuity in his argument
that terrestrial fetiches, because they mostly
could not move about, were sooner detected
and deposed, and became mere material
things; while the celestial fetiches were
very active and could maintain the illusion
better: more myths were therefore told
about them and remained long after the
fetich-belief was exploded ; thus when no
other active and intelligent agency was
admitted except that which was human or
like human, these persistent myths which
seemed to speak of active agency in the
heavens gave rise to the belief in personal
agents existing there, the deities of the
Greek Pantheon.
This exposition is the sanest part of the
book but, as it implies that the personal
divinities were not likely to be evolved from
terrestrial ‘fetiches,’ there are obvious ob-
jections to it: the chief one is the very
primitive belief in the earth-goddess and in
divinities that clustered round her or ema-
nated from her. Gaia—whom M. Ploix
scarcely notices—was a real person with
oracular gifts, who certainly was evolved
from that largest of all terrestrial fetiches,
the earth. And in spite of M. Ploix, divi-
nities such as Cybele, Demeter, Dionysos,
had far more aftinity with Gaia than they
had with twilight.
Another proof he urges in favour of the
exclusive supremacy of the celestial person-
ages is that mankind always worship the
cheerful and benevolent, not the dark and
malign power. Does M. Ploix consider
lightning and snakes as cheerful and bene-
volent, or has he failed to notice that some
old Greeks worshipped Python before Apollo
killed him, and that the Arcadians wor-
shipped unpersonified thunder and lightning:
that Briareus-Aegaeon in Euboea and Tityos,
whom Apollo had to slay for his wickedness,
are not known to have been benevolent
powers, and are yet known to have received
worship? He will not admit the existence
at any time of animal-worship in Greece ;
but if the primitive Greeks never worshipped
animals, they yet exposed themselves to the
gravest suspicions of having done so. Main-
taining then that it is only the celestial phe-
nomena that could give birth to divinities,
he finds that the only department of nature
in which he can conveniently plant his
divinities is the twilight. And certainly the
twilight answers the purpose of explaining
all the Greek deities and heroes quite as
well as the sun; or rather, all the depart-
ments of nature, each in its turn, serve
almost equally well the purpose of physical
allegory.
But the twilight has especial advan-
tages for M. Ploix’ argument: itis beautiful,
and all the deities are beautiful, except
Hephaestos, whom M. Ploix considers to be
a twilight ‘manqué’: it is combative, that
is, the morning twilight prevails over the
powers of darkness, and the gods also are
combative, using spears on occasion, which
are the shafts of the morning light. But
the subtlest proof of all is the proof derived
from the fact that the twilight is the ‘com-
mencement’ of the day, and M. Ploix finds
that nearly all the Greek and Latin deities
were deities of ‘commencement.’ One might
doubt whether one is allowed to style a god
thus, because the worshippers happened to
invoke him when they were beginning some-
thing serious. But M. Ploix can plead this
much in favour of his theory, that most of
the deities and heroes did something and in-
vented something, and therefore must have
‘commenced ’ doing and inventing it.
His observations of nature are innocent,
though irrelevant; but his playing with the
meaning of names is more serious. ᾿Αθάνατος
as a predicate of a god does not mean ‘im-
mortal,’ but ‘not dead,’ as a primitive Greek
might remark of his dawn-deity when he saw
a fresh twilight in the sky. With quite as.
much right the Greek might have called
himself ἀθάνατος when he got up in the
morning. Scarcely any epithets of divinities
will he allow to be moral epithets originally :
bonus optimus he would translate by physical
terms (perhaps bonus: = duonus = double-=
twilight, the twilight being double and
most deities having a double aspect, one
good and one bad: optimus = brilliant, root
‘op’).
He does not see that to suppose of the
distinct Greek and Latin nations that they
could not spontaneously predicate moral
terms of their deities is to suppose them to
be in a state of extraordinary intellectual
savagery.
Then he is abnormally prone to the usual
error of his school of imputing to the ordi-
nary terms of Greek and Latin writers pre-
historic and allusive senses: for instance,
Eros is twilight because among other reasons
Sophocles in his Antigone in the famous ode
calls him ἀνίκατος, and M. Ploix does not
THE CLASSICAL REVIEW. 41
seem to recognise that Sophocles is writing
of the invincible power not of twilight but
of love. He maintains that Ἵμερος, whom
Hesiod places in Aphrodite’s company, does
not mean ‘desire,’ but ‘day’ = ἡμέρα : for he
does not seem to know that the personifica-
tion of abstract ideas is earlier than Hesiod,
and that a capital letter does not radically
change the meaning of a Greek word. (We
might as well regard the ἵμερος ὀαριστύς and
πάρφασις in the girdle of Aphrodite as
twilight-powers. )
Again, he has the audacity to suggest
that ’Avrépws meant the ‘counter-twilight,’
although this is a word and figure that was
invented very late—some thousands of years
after the word Ἔρως could have possibly
meant the twilight. His indifference to
chronology is also shown in his indiscrimi-
nate use of texts, Orphic literature being
quoted as a proof of primeval ideas, and
Albricius’ treatise De deorum imaginibus
being referred to as a classical authority.
Of archaeology he appears to be absolutely
ignorant, and his few references to archaeo-
logical evidence are worthless, But his
worst offence is his philology, which is
perhaps more unscientific than that in which
any modern writer of his school has ever
indulged. ‘Pallas’ and ‘pallidus’ are con-
nected ‘because Greek and Latin are kindred
languages’; Ἔρως is a derivative from the
root ar = ‘brilliant,’ a theory which appears
to be possible only if one disregards the
stern laws of vowel-change, or if one makes
no distinction between a long and a short a.
The mythologic theory and the philology
of M. Ploix are antiquated: and the method
of argument on which the whole book
depends—the deduction of all the deriva-
tive qualities of the divinity from one single
original idea—is wholly unconvincing. He
ignores the fact that a nation as it progres-
ses may predicate new functions and powers
of its divinities independently of the original
idea ; or may adopt from some alien religion
divine personages fully-formed and endowed
with a complex moral nature which, for those
who adopt them, has no connection at all
with any physical phenomena.
L. R: FARNELL.
SOPHOCLES’S GREEK LEXICON.
Greek Lexicon of the Roman and byzantine
Periods from B.c. 146 to a.p. 1100, by
EVANGELENUS APOSTOLIDES SOPHOCLES.
New York, Charles Scribner’s Sons,
1887. £2 2s.
A FULL notice of the above work must
include some account of the life and char-
acter of its author. Professor Sophocles was
born at.Changaranda, near Mount Pelion in
Thessaly, in 1807. He entered the mon-
astery of Mount Sinai as a student and was
connected with it several years during the
Greek revolution (1821-28), living chiefly in
the mother-convent at Cairo. He evidently
cherished pleasant recollections of these early
years, and maintained relations with the
Sinaitic convent to the close of his life. He
came to the United States under the
patronage of a Foreign Missionary Society
(A.B.C.F.M) in 1829, and after studying at
the Academy at Monson, Mass., entered
Amherst College, but did not stay through the
regular course. He taught in Hartford and
was assistant in the Yale College library at
New Haven. In 1842 he became tutor in
Greek in Harvard College and served in that
position until 1845, and from 1847 to 1859.
In 1859 he was appointed adjunct professor
of Greek, but in the next year he was trans-
ferred to the University Professorship of
Ancient, Byzantine, and Modern Greek, in
which chair he remained until his death in
December 1883. He began early to publish
text-books for students of Greek. Most
known among these are his Greek Grammar,
Romaic Grammar, Catalogue of Anomalous
Verbs and History of the Greek Alphabet. His
life at Cambridge was singular and striking
to eccentricity. He never married, and lived
a secluded life, using two rooms for all
purposes, and often acting as his own cook
and house-keeper. He partook of his simple
meals on the very table which served him
as a desk for the writing of this monu-
mental JZewicon. His behaviour towards
strangers, especially his own countrymen,
was not always attractive and he came to
have the reputation among them of being
crabbed and cynical, but those who had the
privilege of a more intimate acquaintance
with him found him cordial and tender-
hearted. His characteristic as a scholar was
penetration. He seemed to see at a glance
42 THE CLASSICAL REVIEW.
what cost others time and labour in ac-
quiring, and so, although he was not versed
in German and not acquainted with the
results of German research in Philology
except as reproduced in English and Latin,
the extent and accuracy of his scholarship
was not inferior to that of the German
masters themselves. The surest evidence of
this is the work before us.
This work was first published as vol. VII.
(new series) of the ‘ Memoirs of the American
Academy’ under the title of ‘Glossary of
Later and Byzantine Greek.’ In 1870 it
was republished by the author in an enlarged
form under its present title, and again in
1887 by the President and Fellows of
Harvard College in a memorial edition. Of
this edition, a note by the editor tells us
that it differs from the second impression
of 1870 by the correction of nearly two
hundred evident inadvertencies and ‘a few
instances of additions.’ The Lexicon is
then substantially as Sophocles left it. As
appears from the title itself, it is a work of
a very different character from Du Cange’s
Glossarium mediae et infimae Graecitatis.
First of all, the periods of the Greek
language covered by the two works are not
the same. Du Cange’s period extends
apparently down to his own day (1688).
Sophocles does not attempt to deal with
writings later than 1100 a.p. the date of the
capture of Constantinople by the Crusaders.
Although the starting-point in both is the
same, it dees not appear that Du Cange
makes any use of the LXX and of the New
Testament or other early Christian writers ;
so that altogether the fields of the two works
coincide only for a space of 500 or 600 years.
Sophocles refers to over 550 titles; Du
Cange to nearly 1100, of which however a
large number are titles of brief works
separately mentioned which appear in
Sophocles’ list unified under titles of collected
works of authors. Thus Sophocles’s 550
titles represent by no means a proportion-
ately lesser body of literature. Again, of
Du Cange’s 1100 titles 650 were edited at the
time he wrote his Glossary and the re-
mainder were consulted in MS. form. Since
so large a portion of his field was inaccessible
to scholars generally, Du Cange as a rule
gives complete quotations for illustration of
his definitions, while Sophocles in general
simply refers to passages of published works.
Hence Du Cange saves the student the labour
of searching through whole sections in order
to verify references, but loses in point of com-
pactness and clearness. Sophocles is able to
compress into the same space a larger
number of definitions and references than
Du Cange. We turn to a word like λύσις
and find in Du Cange two senses attached to
it: (1) rescripta et responsu imperatorum,
supported by seven citations, and (2) recon-
ciliatio ecclesiae, supported by one. In
Sophocles the same word is found in nine
different meanings: (1) loosening, with one
reference, and idiom λύσις κοιλίας, (2) divorce,
two reff., (3) absolution, two reff., (4) barley-
meal, two rett., (5) resolution of a long syllable
to two short ones, two reff., (6) omission of
conjunction = asyndeton, one ref., (7) rescrip-
tum, four reff., (8) seminal emission, one ret.,
(9) permission, leave, one ref. The common
word ἀφορίζω is not found in Du Cange, but
its radical meaning is given under the deriv-
ative ἀφορισμός as censurae ecclesiasticae
species communionis privatio, with three dis-
tinct citations and a passim reference to the
Synodical Decrees of the Greek Church.
Sophocles in treating the word gives the verb
separately with the following senses, each
abundantly supported by references: (1) to
separate, (2) to separate and cut out, (3) to
excommunicate = ἀκοινώνητον ποιεῖν, (4) to put
in the ἀφορίστρια (a sort of dungeon in
monasteries); and in addition he treats
separately of the following derivatives,
altogether omitted by Du Cange: ἀφόρισμα,
ἀφοριστικός, ἀφοριστικῶς, ἀφορίστρια. In the
treatment of this latter word there appears
another feature of Sophocles’s work as com-
pared with that of Du Cange, viz. the
methodical arrangement of the various
definitions and a general progress from the
generic to the specific, and from the etymo-
logical to the historical sense. Perhaps
another common word of the Byzantine
period will illustrate this difference. For
σήκρητον (σέκρετον) Du Cange gives the
meanings: (1) secretariwm aedis sacrae, (2)
secretarium forum tribunal juridicum.
Sophocles is more methodical as follows :
(1) secret, (2) privy chamber, (3) office, apart-
ment of an office, (4) an association (Porph.
Cer. 6). Very notable is this difference in the
word εὐαγγέλιον. Both lexicographers give
it eight different senses, but an examination
shows that those given by Du Cange are
within a narrow range, whereas Sophocles’s
treatment proceeds on a regular method and
covers all the usages of the word, as the
following summary of the treatment will
make plain. Du Cange: (1) evangelium,
Novum Christi Testamentum, (2) liber ecclesiae
Graecorum in quo descripta sunt evangelia,
(3) = εὐαγγέλια ἐωθινά, (4) = εὐαγγέλια ἐωθινὰ
ἀναστάσιμα, (5) = εὐαγγέλια ὁσιακά, (0) =
μικρὰ εὐαγγέλια, a mulieribus vice φυλακτηρίων
THE CLASSICAL REVIEW. 45
gestata, (7) = εὐαγγελιστάριον, (8) = τετρα-
ευάγγελον, codex quatuor evangehorum, Xe.
Sophocles: (1) good tidings, (2) applied to
the revelation by Christ = Gospel, (3) Gospel,
a history of Christ, (4) the gospels collectively,
(5) the book containing the four gospels, (6)
the gospel of the day, (7) evangelistary, the
book containing the gospel of each day, (8)
a book of faith; cf. εὐαγγέλιον Evas, ᾿Ιούδα,
Σκυθιανοῦ. This compactness and method
naturally enable and lead Sophocles to
notice unusual meanings of words not noted
by Du Cange. The word ἐκκλησία is found
in Du Cange in the ordinary senses of : (1)
conventus Christianorum, (2) Aedes in quam
conveniunt fideles, and (3) cathedrales civitatum
(cf. μεγάλαι ἐκκλησίαι). Inaddition to these,
all mentioned by him, Sophocles gives: (4)
local church, (5) Ecclesia, female of Aeon in
the Valentinian theogony, (6) Roman comitia,
and a note on of ἀπὸ τῆς ἐκκλησίας ΟΥ οἱ τῆς
ἐκκλησίας = members of the true church.
Quite often the two works are equally full,
but each in a different line ; e.g. εὐλογία is
given in Du Cange in nine senses and in
Sophocles in seven. Of Du Cange’s nine the
first appears as (3) in Sophocles, the second
as (4), the third as (5), the fourth and fifth
not at all, unless they be referred to (1) =
blessing in general ; the sixth appears as (7),
the seventh as (2); the eighth, like the
fourth or fifth, not at all or under (1), and
the ninth (pestis species—small pox, no
doubt) not at all. On the other hand,
Sophocles adds as (6) = nuptial benediction
= στεφάνωμα. Similars the case with other
words, cf. σπαθάριος, ἀββᾶς, τόπιον. Occasion-
ally a word quite fully treated in
Sophocles is not found in Du Cange’s
Glossary, cf. ὀφείλω, σαλεύω. But above
these differences, a special excellence of
Sophocles in comparison with Du Cange is
his philological treatment of words. His
knowledge of the history of the. Greek
language is complete; he is familiar not
merely with classic antecedents of his
special period of that history, but also with
the various streams that have flowed in and
mingled with it, and as a rule he traces words
of foreign origin to their respective sources,
as he also generally connects purely Greek
words with their classic antecedents. Thus
Latin words are traced to the Latin, and
Hebrew or other Shemitic words to their
Shemitic roots. There are of course ex-
ceptions, cf. ἀββᾶς, Φαρισαῖος, Σαδδουκαῖος ; but
perhaps these are treated like proper names.
The Egyptian origin of Batov (as per
Stephens) Sophocles either consciously or
unconsciously omits. Combined with this
linguistic knowledge we discern a peculiar
familiarity with the customs, institutions,
and history of the Eastern (Greek) church
in Sophocles which at times goes beyond
even Du Cange’s thorough acquaintance
with these things. On δεσπότης, for example,
Du Cange merely gives the secular sense of
dominus ; while Sophocles gives: (1) master
in general, (2) emperor, emperor's son, and (3)
the ecclesiastical bishop, priest.
This comparison need not be carried into
matters of external and mechanical execu-
tion, such astypeand form. It goes without
saying that in these respects the more
modern work is an improvement upon the
old. It is enough to point out the cumbrous
system of appendices and addenda in Du
Cange, in contrast with Sophocles’s perfect
simplicity. Nor need anything more than
the mere fact of Sophocles’s independence
from Du Cange be noted here. His method
of treatment, his point of view, his references
and quotations are different. A comparison
of ten words with an aggregate of about
150 references by Du Cange and 200 by
Sophocles shows that only about 15 to 20
references are identical. (Greater precision is
impossible owing to the difference in the
titles as quoted by the two lexicographers,
and the inaccessibility of many of Du
Cange’s authors.)
Altogether, Du Cange’s Glossariwm 15
more of a collection of short historical
monograms on a list of words, whereas
Sophocles’s Zeatcon is a lexicon with more
or less precise etymological and methodical
definitions.
But aside from comparison with Du
Cange’s Glossary, this work of Sophocles
may be commended as at present the only
aid to the scholar in his studies of the earlier
Greek Fathers. It is not perfect of course.
In its introductory portion, one or two
changes would increase its scientific and
logical character: e.g. in the periodology of the
Greek language, we fail to see the import-
ance of the hegiric era 622 as an epoch-
making date; the event it marks was too
remote to exercise a direct influence on the
course of the Greek language. It would be
much better to substitute the year 780 (the
date of the separation of the Eastern and
Western churches), as marking a new de-
parture in Byzantine history. A similar
change of the beginning of the Alexandrian
period from 283 to 332, though less obvious,
would nevertheless be a real improvement.
Every one will sympathise with the author
when he tells us in the prefatory note that
he had been compelled to leave out certain
44 THE CLASSICAL REVIEW.
classes of words there described, ‘ the alter-
native being to give up the intention of
publishing the book.’ AJl omissions of words
familiar to the scholar from classical usage
will naturally be attributed to this same
necessity ; but why should distinctly post-
classical (Roman and Byzantine) usages be
omitted, such as the meanings of κύων (1) =
Jetlock joint of a horse (Hesych.), (2) = κυνικὸς
σπασμός (Gal. VIII. 41), (3) = crust used as
spoon and thrown to the dogs (Hust. 1859,
19), (4) =a throw at dice; Lat. canicula
damnosi canes (Poll. 1X. 100; Eust. 1289) ?
Or under διδάσκω the phrase διδάσκειν τὸν
λαόν = to preach, or under κλέπτω the sense
of concealing (Syn. 283 C) ?
Again Sophocles is not perfectly consistent
in the omission of classical words; or in
the grammatical suggestions appended, e.g.
he gives the fut. of βηματίζω but the aor. of
βιγλεύω, βιαιοθανατέω, and neither the fut. nor
the aor. of βιαιομαχέω. Or, in proper names,
he gives Ἱεροσόλυμα, Ἱεριχώ, ᾿Αλεξάνδρεια, but
leaves out Δαμασκός, Ἰόππη &e., justascommon
and important, or Tdyada, Γίσχαλα, ᾿Ιωτάπατα
&e. (Josephus) less known. If the choice in
all these points is on principle, a brief intro-
ductory note would help the user as a guide
when to consult and when not to consult
this Leaicon. It would have been better
also, especially for the use of students, to
give the English of Greek equivalents, either
with or instead of the Greek. The word
ἀβίβλαβον is given as = κρίνον, ἡμεροκαλλίς,
which is familiar to the scholar, but not to
the student ; and so with a large number of
equivalents.
But these are mere blemishes in a great
work. It is not to be denied that a great
step forward was taken in Greek lexico-
graphy when this Lexicon made its appear-
ance. But the diligence with which _his-
torians are at present investigating Byzan-
tine literature deserves and demands a fuller
and more scientific work, and the next step
forward must be in that line. Hence we
cannot look with unmixed joy upon the
announcement of a facsimile reprint of Du
Cange’s Glossary. If this reprint is to serve
as a mere historical monument, it is well ;
but if it is to occupy the place which be-
longs to a more modern and exhaustive
work, it is to be deprecated.
A. C. ZENos.
Volistandiges Worterbuch uber die Gedichte
des Homeros und der Homeriden, nach dem
fritheren Seilerschen Homerworterbuch nen bear-
beitet von CAPELLE. Neunte verbesserte Auflage.
Leipzig, 1889. Mk. 4.80.
THE ninth edition of a well-known work, improved
but not greatly altered, demands no extended review ;
but the preface calls attention to the notable number
of important works on Homer which have appeared
since the publication of the eighth edition in 1889,
and every page bears the marks of careful revision.
In the great majority of cases, the best meanings and
derivations are given concisely, and a long list might
be made of words about which this dictionary gives
better information than is found in the seventh edi-
tion of Liddell and Scott’s Lexicon. 1.5. πεφυζότες
X 1 must mean ‘frightened,’ cf. φύζα 1 2, for which
L. and S. makes no provision; μέλλεν μέν ποτε
οἶκος ὅδ᾽ ἀφνειὸς καὶ ἀμύμων a 232, L. and S. ‘was
destined to be’ etc., instead of ‘doubtless was,’ as
Capelle rightly has it. In αἴ κεν ἐμοὶ Ζεὺς | δώῃ καμ-
μονίην, σὴν δὲ ψυχὴν ἀφέλωμαι X 256 f., L. and S.,’
following a prosaic scholiast, renders καμμονίην by
“aictory as the reward of steady cowrage.’ Far better,
if not necessary, is it to take καμμονίην literally (with
Capelle, ‘Ausdauer’), ‘steadfastness.’ ‘If Zeus
grants to me to stand steadfast, and I take from thee
thy life.’ Here the first half verse is explained by
the second.
Occasionally Capelle is unsatisfactory, and guilty
of such slight inconsistencies as naturally arise in
such a work. £g. under ψευδής he treats of ψευδέσσι
A 235; but explains this as ψεύδεσσι, under ἀρωγός.
But an immense amount of accurate information is
contained in this book of 605 pages, sold for about
four shillings. This dictionary is ἃ condensed com-
mentary, and if a man were limited to one book on
Homer, besides the text, he might well choose this.
ΠΕΡΉΣΘΣ
Dinarchi orationes adjectis Demadis qui fertur
fragmentis ὑπὲρ τῆς δωδεκαετίας, iterum edidit
FRIDERICUS BLAss. 1888, Teubner. 1 Mk.
SINCE 1865 Professor Blass has been active as a
student of Greek orators and oratory and has acquired
a well deserved prestige in this field. Seventeen
years have intervened between the first and this the
second edition of Dinarchus. It would indeed be a
matter of substantial interest to know how many
copies of the first edition were issued, for it would
enable one to judge of the extent of classical reading
at large, particularly in the case of this, the last of
the ten Attic orators, both in time and in literary
attractiveness. One may say that these slender be-
quests of the Corinthian speech-writer owe their
survival to the interest which history has shown in all
the literary tradition that bears upon the career of
Demosthenes,
The two most valuable MSS. are both in England,
an Oxoniensis and a Crippsianus.
On the whole Blass has been conservative as an
editor, rarely on the whole admitting the inter-
polations claimed by the Cobetian critic Kleyne,
Mnemosyne (older series), viii. Blass says very
sensibly, praef. p. ix. : ‘ ipse id tenwi ne quod verbum
ideo delerem, quia abesse potest vel posse videtur.
Nimis facile cst in quovis scriptore invenire quae
recidere possis ; sed non est oratoris qualis hic est
quan paucrssimis ubique verbis uti.’ Indeed one may
THE CLASSICAL REVIEW. 45
fitly characterize this speech-writer as long-winded
and strong winded. It may safely be doubted
whether in all the compass of classic Greek prose
there can be found a more ponderous or more awk-
ward period than that beginning §18 contra Dem. οὐκ
ἀποκτενεῖτε and ending μὴ βοηθῆσαι τοῖς Θηβαίοις at
the conclusion of § 21 (οἴ, Transactions of the
American Philol. Asseciation for 1885).
Some of the points to be noted after a comparison
of the first and second editions of Blass’
Dinarchus are the following. Blass now proceeds
radically in making elision to avoid hiatus, not hesi-
tating even where the resulting form might be ambig-
uous, as μηδέν᾽ for μηδένα. Many changes have been
made in the matter of pointing, generally in the
direction of mere articulation. In questions of form
and spelling he desires conformity with epigraphic
evidence of the times and refers to Meisterhans’
Grammatik der Attischen Inschriften. He writes
the perf. of τρέπω τέτροφα. With one exception (§54
c. Dem.) he writes γίγνεται, not γίνεται, and spells
Μουνιχία and ἐκτείνω for ἐκτίνω, ἐκτείσις for ἐκτίσις,
probably from epigraphic considerations. Of passages
where the reading is still in grave doubt we may note
§ 2c. Dem. ov yap tv’ ἐνοχλῶμεν κιτ.λ. Following
Meisterhans he reads ἐπιτεθηκότες in ὃ 23 c. Dem. tor
ἐπιτεθεικότες of the MSS.
A few critical suggestions may be here appended.
It was suggested (Proceedings Am. Philol. Assoc. for
1885, and Am. Journal of Philol. vol ix.) to insert
οὐδὲν after πώποτε c. Aristog. 15, in which remark we
coincide with Bekker, τοιοῦτον for τοῖς τοῦτον in the
same section-would seem to be more in harmony with
grammar and context. E. G. SIHLER.
New York.
In the praefatio xi. line 4, Hermes xxii. 78 is a
mistake for 378.
Supplementa ad Procli Commentarios in Pla-
tonis de republica libros nuper uulgatos,
edidit RicARDUS REITZENSTEIN. Breslauer Philo
logische Abhandlungen. Vierter Band. Drittes
Heft. Breslau, 1889. pp. 1—31. 1 Mk.
Tus short pamphlet has been issued in view of
Cardinal Pitra’s recent publication (from the MS.
Gk. Vat. 2197) of Proclus’s commentary on the story
of Er, the son of Armenius, in Tom. v. part ii. of the
Analecta Sacra Spicilegio Solesmensi parata (Parisiis
1888). Cardinal Pitra seems to have been unaware
that a second copy of this work of Proclus, made by
Angelus Maius, is preserved in the Vatican library :
and Reitzenstein has published a revised text of part
of the commentary (= pp. 25—32 and pp. 1—12 of
Vat. MS. 2197) from a comparison of the original
MS. with Maius’scopy. Pitra’s text contained many
errors, and probably was not finally revised by him
when he died. Reitzenstein has corrected most if not
allof his mistakes, and made many certain emendations
where the manuscript is unintelligible, e.g. ὁ αὐτός for
οὗτος on page 7, line 2: μὲν ὄν tor μένον on 7, 25:
δίκην for διπλῆν on 9, 15: ἀνάγειν eis τὸν αἰθέρα for
ἀνάγει τὸν αἰθέρα on 10, 24: ποιεῖν ἔοικεν for torn κεν
of MSS. in 12, 14: χάσμα for χίασμα on 17, 80 : ar
ἄλλων for am ἀλλήλων on 29, 14: and ὄχημα for
σχῆμα on 31,7. It may however be doubted if the
editor is justified in inserting ἐπὶ before τὸν αἰθέρα on
page 10, 26, ὡς ἐπὶ δικαστικὸν τόπον ἀνιέναι τὸν
ἀιθέρα : in such cases, as is well known (Cohet, Var.
Lectiones, p. 165 foll.), the preposition is idiomatically
inserted only once, viz. after the ὡς (ὥσπερ, καθάπερ),
provided the ὡς clause comes first. We have no
reason for denying this elegancy even to Proclus.
Reitzenstein will confer a boon on readers of Plato
and Proclus if he will publish the remaining parts of
the commentary. J. ADAM.
Bellum Alexandrinum, erklirt von R. SCHNEIDER.
Berlin: Weidmann. 1888. pp. viii. 65. 90 pfg.
THE writings of Caesar’s continuers were, until lately,
among the most neglected portions of Latin literature.
The text had been edited but textual criticism had not
been applied in detail and, since 1830, no commen-
tary appears to have been written on any part of them
except the last book of the ‘Gallic War.’ But theenergy
recently developed in the study of Caesar has extended
to his continuers also. Merguet and Menge-Preuss
have indexed them, and Dr. Schneider offers a brief
commentary on the pamphlet describing the Alexan-
drine war, that is the events from the battle of Phar-
salus to Caesar’s entry into Rome. The book com-
prises a short introduction, critical and explanatory
notes, and an index. The critical notes give the
readings of the four chief MSS. and select conjectures.
In constituting the text, Dr. Schneider has been
reasonably conservative (as conservatism is now
understood); he appears, however, to rate the
literary capacities of his author higher than previous
scholars have done. Some of his conjectures are
certainly attractive, e.g. ch. 8 consolativne et hor-
tatione for ὁ. et ratione ; ch. 17 Pharitae for pariter ;
ch. 42 in Illyricwm sinwm for in illum s., and the
geographical restorations (for instance in ch. 52,
lipa and Naeva) are well worth considering. If I
may venture myself a suggestion, I would suggest
appetuntur for aptantur in ch. i., comparing Caesar’s
appetere umerum gladio (Bell. Civ. 2, 35). On the
whole, Dr. Schneider's text is a distinct improvement
on previous editions. The same cannot be said of the
notes—for there is practically no previous edition on
which they could constitute an advance. But they
are (as would be expected from the writer) well chosen
and scholarly, and, though never lengthy, they
seldom leave the reader in the lurch. The whole
book deserves to be very warmly recommended.
ΒΕ, HAVERFIELD,
1 des
Berlin,
Der Codex Mediceus, Pl. xxxix. ἢ.
Vergilius, von Dr. Max HoFrMAnn.
Weidmann, 1889. 3 Mk.
THIs is a new collation of parts of the Medicean
manuscript of Vergil, comprising part of the sixth
Eclogue, and the remaining Kclogues, the Georgics,
and the first and sixth Aeneids. In minute details
it supplements Ribbeck, who mainly depended on
Foggini’s transcript. ‘lhe account of the different
classes of corrections is especially interesting. As
the corrected or double readings are all printed in
facsimile, the collation will be of great service to stu-
dents of Latin palaeography ; but whether it throws
much new light on the text of Vergil is very doubt-
ful. HN
Lettres Inédites de Michel Apostolis, publices
par HipeoLtytTEe Norrer. Paris, Thorin, 1889.
A very interesting and valuable contribution to
the history of Greek erudition in the fifteenth cen-
tury. Michel Apostolis was a professor in Crete from
1455 to 1474 (or later) and lived during this troubled
time by teaching, and copying Greek manuscripts.
Forty-eight of his letters (the forty-seventh being
incomplete) had been published by Legrand in the
appendix to the second volume of his Bibliographie
Heéllenique. The present volume contaius seventy-
four more letters of Apostolis, hitherto unpublished,
and now edited by M. Hippolyte Noiret, whose early
death at the age of twenty-four is a great loss to letters.
They were found in two Vatican MSS., Palatinw:
Graecus 875 (once at Heidelberg) and Vaticanws
40 THE CLASSICAL REVIEW.
Graecus 1395, ‘The volume was prepared for the
press and brought out by M. Desrousseaux, to whom
Noiret was indebted for his first acquaintance with
the two Vatican manuscripts. It contains (1) an
account of the manuscripts: (2) an essay on the clas-
sification and chronology of the letters: (3) a bio-
graphical notice of Michel Apostolis and his numerous
correspondents : (4) the letters themselves. In the
case of the letters previously edited by Legrand only
a collation of the newly discovered manuscript is
given ; the rest are printed in full.
HN:
The Cradle of the Aryans. By G. M. RENDALL.
London : Meecmillan. 1889. 95.
Tuts little work is not, and does not pretend to be,
more than ‘a clear account of present controversies
regarding the origin, local and racial, of the Aryan
stock.’ ‘The work is done probably as well asit could
be done in the limited space of 63 pages. Mr Rendall
gives the arguments of Schrader against accepting
the Asiatic hypothesis, summarises Penka’s works,
and concludes in favour of ‘the large northern
plateau extending from the German Ocean to the
contines of the Black Sea and Caspian watersheds’
as the local origin of the Aryans ; and considers that
‘ Penka has gone far towards establishing an associa-
tion between Aryan speech and the race of blond
whites, whose central and immemorial home is found
in Scandinavia.’
Mr. Rendall ‘makes no claim to original dis-
coveries in the field of language.’ In the field of
anthropology he bases himself mainly on Penka, with
the proviso that he is ‘not competent to check
Penka’s anthropology.’ However, if the English
reader (for whom the book is intended apparently)
comes, as he probably will come, to the conclusion
that Penka’s speculations are as yet too remote from
actual facts, philological and anthropological, to pro-
vide even a working hypothesis, that will not be the
fault of Mr. Rendall’s way of summarising Penka.
From the point of view of the history of culture one
or two observations may be made. Mr. Rendall ap-
proves of the beech argument. But at the recent
meeting of the British Association doubts were ex-
pressed as to the Konigsberg-Crimea line: further
search in the bogs east of that line might very well
reveal the existence of beeches. Still greater weight
is assigned by Mr. Rendall to the anguilla—éy-
xedus—ungurys—aguija equation for ‘eel.’ He says:
‘here then at last we are driven to a corner, and
must accept the inference that the European unity
was maintained...... ut some point in Central or West-
ern Europe which the evidence forbids us to place far
east of Russia’s western frontier. This if tenable will
prove decisive to the whole controversy.’ But it is
at least conceivable that the agreement between the
languages cited is casual, that each people quite inde-
pendently of the rest used for the designation of the
eel, when they made its acquaintance, a diminutive
of a pre-existing common word for ‘snake.’ Surely two
or more peoples might quite independently have
called the eel a little snake? The lion question
again is not quite settled by Mr Rendail. He may
be quite right in concluding that the primitive Indo-
Europeans knew nothing of the lion; but the fact
that all members of the European branch have a
common name for it—even though a loan-name—
would seem to indicate that they jointly made the
acquaintance of the lion after leaving the original
home. The Indo-Iranians again made the lion’s ac-
quaintance separately after leaving the original home,
wherever it may have been. Obviously we want
some place which did not rejoice itself in lions, but
was bounded both on the east and on the west by
districts more favoured in this respect. Such a
locality would be (not W. of the Konigsberg-Crimea
line but) the southern steppes of Russia, supposing
the distribution of animals was in pro-ethnic times
the same as it was in the times of Herodotus and
Aristotle, who testify to the presence of lions in
Thrace and the neighbourhood. Again, assuming
that J. Schmidt’s Undulation or Transition theory is
true, and admitting that we must therefore suppose
the original home to have been free from great physi-
cal obstructions, such as the Caspian and the Cau-
casus, we can only infer from this that the original
home was a plain or plateau; we are not compelled to
assume that the plain was west of the Caspian and
north of the Caucasus. Finally, Mr. Rendall uses,
I think, ‘ Aryan’ indiscriminately as equivalent both
to Indo-Germanic and Indo-Iranian ; and this must
be puzzling, ¢.g. on p. 34, to the class of reader for
whom the book is intended. Schmidt’s Verwandt-
schaftsverhdlinisse d. Indog. Spr. more than once
appears as Verwantschaftsverhalimisse etc.
F. JEVONS.
NOTES.
Hom. Ji. xix. 221 sq.
αἶψά τε φυλόπιδος πέλεται κόρος ἀνθρώποισιν,
fs τε πλείστην μὲν καλάμην χθονὶ χαλκὸς ἔχευεν,
ἄμητος δ᾽ ὀλίγιστος, ἐπὴν κλίνῃσι τάλαντα
Ζεὺς, ὅς τ᾽ ἀνθρώπων ταμίης πολέμοιο τέτυκται.
225, γαστέρι δ᾽ οὔ πως ἔστι νέκυν πενθῆσαι ᾿ΑχαιούΞ"
λίην γὰρ πολλοὶ καὶ ἐπήτριμοι ἤματα πάντα
πίπτουσιν: πότε κεν TLS ἀναπνεύσειε πόνοιο ;
‘Swiftly, we know, do men find surfeit of battle,
wherein, though the steel sheddeth on the earth
haulm in plenty, yet the harvesting is but scant,
what time Zeus hath inclined the balance, Zeus, who
is a dealer out of war for men.’ Odusseus had
already (157) given his vote against sending the men
fasting to the field, ἐπεὶ οὐκ ὀλίγον χρόνον ἔσται
φύλοπις : he is here only repeating that advice under
a metaphor. ‘The men,’ he says, ‘have a long
(πλείστην) and thankless (oAljioros) day’s work
before them: therefore they should have a good
meal first, for they will find that Zeus is a ταμίης
πολέμοιο, ποῦ a Tauins, σιτοῖο δοτήρ (44). ἐπὴν
κλίνῃσι K.T.A. Seems to mean no more than ‘ when
the day is over,’ the aor. subj. having, as is usual in
this form of sentence, the force of an English fut. per-
fect. φύλοπις, then, will not be ‘a pitched battle’ as
opposed to a ‘siege,’ nor will ἄμητος have any refer-
ence to ‘plunder,’ but is merely due to the meta-
phorical cast of the thought, which is kept up in
226, ‘for very many in number and in thick piles
are they mowed down,’ as in 18, 533: δράγματα δ᾽
ἄλλα μετ᾽ ὀγμὸν ἐπήτριμα πίπτον ἔραζε (quoted by La
Roche, who notes that πίπτουσι δ passive ot ἔχευε, 221).
And this perhaps determines the sense of the next
THE CLASSICAL REVIEW. 47
words (227), which Mr. Monro seems to regard as a
merely parenthetical exclamation equivalent to ‘we
are never done fighting.’ But if this were the sense,
it may be doubted whether Homer would have added
κεν, Which is quite in place if the words have their
protasis in the context: ‘if we are to fast for every
man that falls, then (κεν) indeed we have our work
(πόνοιο) cut out for us.’ That this sense of πόνος,
‘hardship of fasting,’ is not ‘ post-Homeric’ perhaps
appears from 22, 488 :
iy περ yap TéAE MOV γε φύγῃ πολύδακρυν ᾿Αχαιῶν,
αἰεί τοι τούτῳ ye TOVOS καὶ κήδε᾽ ὀπίσσω
ἔσσοντ'᾽ ἄλλοι γάρ οἱ πουρίσσουσιν ἀρούρα-.
Hom. Ji. xix. 76.
τοῖσι δὲ καὶ μετέειπεν ἄναξ ἀνδρῶν ᾿Αγαμέμνων
αὐτόθεν ἐὲ ἕδρης, οὐδ᾽ ἐν μέσσοισιν ἀναστάς.
‘Agamemnon spake from the place where he had
been sitting and not after going to stand in the
midst of the assembly.’ Sch.: ot προελθὼν εἰς
μέσην τὴν ἐκκλησίαν, tv’ ἦ, ἀναστὰς ἐδημηγόρει οὐκ ἐν
μέσοις, ἀλλ᾽ ἐκ τῆς ἕδρας τῶν βασιλέων. The above
view does not appear to conflict with Od. 18, 56:
οἱ δὲ θεοῖσιν
4 Uj > FN ¥
ἔσπεισαν μακάρεσσι, Tol οὐρανὸν εὐρὺν ἔχουσι,
αὐτόθεν ἐὲ ἑδρέων' ἀνὰ δ᾽ ἵστατο δῖος ᾽Οδυσσεὺς
2 L aes / , > U
Αρήτῃ & ev χερσὶ τίθει δέπας ἀμφικύπελλον---
for surely the Phaeacians did not ‘say grace’ sitting,
and ἀνὰ δ᾽ ἵστατο obviously implies that Odusseus not
merely rose from his seat but moved away to another
place. L. 77 is bracketed by Mr. Monro and by Mr.
Leaf, who objects that ‘the distinction (between ἕδρης
and ἐν μέσσοισι) is void of meaning.’ But his own
note on 85 might suggest a defence of the line. It
may be a part of the poet’s attempt to convey the
embarrassment Mr. Leaf well points out in Agamem-
non’s opening words, natural to a speaker who knows
heis to blame (86 ἐγὼ δ᾽ οὐκ αἴτιος εἰμί) and piqued at
the applause (81 πολλῷ ὁμάδῳ) which had followed
Achilleus’s gallant speech (56-73). Already in 51 we
find Agamemnon coming to the assembly δεύτατος,
ἕλκος ἔχων, making as it seems more fuss about his
wound, though it was after all only in the arm, than
did Tydides and Odusseus about theirs, who though
hit in the leg arrived before him, σκάζοντε, and took
their places μετὰ πρώτῃ ἀγορῇ (47-50). Where
Agamemnon sat we are not precisely told ; but it
seems that on arising to address the assembly he was
shy of taking the prominent place usually occupied
by a speaker in the ἀγορά. If then the line is an
‘emblema,’ our old friend the interpolator cannot
have been such a bad fellow after all: at least his
verse is not unworthy of the author of a similar,
though finer, touch in 21, 36, where Achilleus comes
an ἀνώιστον κακὸν to young Lykaon:
6 δ᾽ ἐρινεὸν ὀξέι χαλκῷ
τάμνε νεοὺς ὕρπηκας, iv’ ἅρματος ἄντυγες εἶεν.
But we now see that Mr. Leaf, with a far more
‘remorseless hook’ than innocent Lykaon’s, threatens
to lop off this beauty too trom Homer’s book.
W. T. LENDRUM.
* *
*
HecatTe.—Mr. Bury’s ingenious etymology seems
to be based on a fallacy. The comparison of Ἑκάτη
with ἑκατόν is specious, but surely some account must
be taken of the first syllables. It is not ἑκατόν
but *-xaroy that corresponds to the Goth. hund.
There seems little objection to the explanation of the
é- adopted by Brugmann Gr. Gr. ἃ 101. In any
case it must be either (1) formative—has Mr. Bury
any similar explanation to offer for the corresponding
syllable of ‘Exarn ?—or (2) an (impossible) accretion—
does Mr. Bury regard the ‘E- as a similar accretion
to a different root, and if so can he justify such an
extraordinary coincidence? The same objection will
apply to his explanation_of éxdFepyos and ἑκηβόλος
as formed from the base &(w)2.—
Mr. Bury makes Hecatus a masc. formed from
Hecate, rather than wice versa. Is it not of some
evidential value that the epithet ἕκατος of Apollo is
common in Hom., whereas Ἑκάτη is post-Homeric ?
ἑκατηβόλος may very Well stand as a contamination
form from ἕκατος and ἑκηβόλος.
Finally a hint of at least the Homeric interpreta-
tion of the epithets may perhaps be seen in JZ. 1. 48.
J. H. VINCE.
* *
ΕἾ
A PINDARIC SCHOLION. —A vetus scholiwm on Pin-
dar, Nemcean iii. 62 ἐν φρασὶ πάξαιθ᾽ runs thus (ed.
Abel, p. 103) :
πλαγίω 5 λογίσαιτο καὶ κρίνοι" ἀντὶ τοῦ εἰς πέρας ἄγοι
Abel’s note is: num μή ante πλαγίως addendwm? I
am surprised that none of the many German scholars
who have investigated the Pindaric scholia vetera
perceived the obvious correction of πλαγίως.
Read taylws λογίσαιτο κ.τ.λ.
The whole point of the explanation of πήξαιτο is
contained in παγίως. ---
Jee UR
* Ἀ
*
ΞΟΡΗ. Antigone, 2—3.
ap’ οἷσθ᾽ ὁ τι Ζεὺς τῶν am’ Οἰδίπου κακῶν
ὁποῖον οὐχὶ νῷν ἔτι ζώσαιν τελεῖ;
It seems to be agreed that either ἐστὶ or τελεῖ is to
be supplied here—with 6 7: most editors say. Is it
not possible that the “τελεῖ in the text is the “τελεῖ᾽
which belongs to 6 τι and that it is with the other
clause that τελεῖ has to be ‘supplied’ or ‘under-
stood’? If Sophocles had written
ε
ἄρ᾽ οἷσθ᾽ 6 τι Ζεὺς τῶν ἀπ᾿ Οἰδίπου τελεῖ,
ὁποῖον οὐχὶ νῷν ἔτι ζώσαιν, κακῶν ;
he would have written harshly, but the construction
would have been plain : the parenthetic clause would
mean ‘ otherwise than upon us in our lives.’ Now if
the clause ὁποῖον... ζώσαιν be isolated by commas, or
by pauses in actual delivery, the construction is
equally plain :
ap’ οἷσθ᾽ 6 τι Ζεὺς τῶν ἀπ᾽ Οἰδίπου κακῶν,
ὁποῖον οὐχὶ νῷν ἔτι ζώσαιν, τελεῖ ;
This is not altogether satisfactory, but the construc-
tion would at least be fairly clear to a hearer. How
could an actor deliver the first line in such a way as
not to excite vain expectation of a verb to follow ὃ
This he would have to do, if the τελεῖ in the text
belongs to ὁποῖον ov. W. R. HArpieg.
% 8
*
ΞΌΡΗ. Anfé. 1. 4.
οὐδὲν γὰρ οὔτ᾽ ἀλγεινὸν οὔτ᾽ ἄτης ἄτερ
οὔτ᾽ αἰσχρὸν οὔτ᾽ ἄτιμόν ἐσθ᾽, ὁποῖον οὐ
τῶν σῶν τε κἀμῶν οὐκ ὕπωπ᾽ ἐγὼ κακῶν.
This, the reading of the MSS., yields no tolerable
sense, and most editors have directed their emenda-
48 THE CLASSICAL REVIEW.
tions to the words οὔτ᾽ ἄτης ἄτερ. Prof. Jebb,
however, has drawn attention to the scholium in ἢ
which proves that ἄτης ἄτερ was already the tra-
ditional reading in the time of Didymus, circ. 30 B.c.
The Scholium runs: Δίδυμος φησὶν ὅτι ἐν τούτοις τὸ
ἄτης ἄτερ ἐναντίως συντέτακται τοῖς συμφραζομένοι"5"
λέγει γὰρ οὕτως: οὐδὲν γάρ ἐστιν οὔτε ἀλγεινόν, οὔτε
ἀτηρόν, οὔτε αἰσχρὸν ὃ οὐκ ἔχομεν ἡμεῖς. ἄτης ἄτερ δέ
ἐστι τὺ ἀγαθόν. Is it ποῦ possible that the true corrup-
tion may be confined to the word ἀλγεινὸν, and that the
line may originally have stood thus: οὐδὲν yap οὔτ᾽
atnpodv οὔτ᾽ ἄτης ἄτερ In this case we shall have both
in ἀτηρὸν and οὔτ᾽ ἄτης ἄτερ a reference to the semi-
technical sense in which ἄτη is used by the trage-
dians of an hereditary curse, so that we may para-
phrase: ‘There is no misfortune involved by the
curse upon our house, nor any general calamity
either (i.e. not directly referable to that curse), etc.
It would be in this sense then that Didymus
remarked τὸ ἄτης ἄτερ ἐναντίως συντέτακται τοῖς
συμφραζομένοις. (The rest of the scholium I take to
be not the words of Didymus, but an attempted
explanation of them by the scholiast in Z. The reading
ἀλγεινὸν may have originated from a gloss on the
rarer word ἀτηρὸν by a scribe who, failing to grasp
the technical sense of ἄτῃ, did not perceive the true
antithesis of arnpdy and ἄτης ἄτερ.) C285
* *
ΕἼ
ΞΌΡΗ. Phil. 1383.
πῶς yap τις αἰσχύνοιτ᾽ ἂν ὠφελούμενος ;
This line is usually left with the manuscript
reading on the supposition that the verb has a unique
middle sense. Wecklein tries to avoid this by sug-
gesting ὠφελῶν τινά. If we read simply ὦ φίλ᾽,
ὠφελῶν, it is a perfectly easy matter to trace the
process by which the manuscript reading was evolved.
The very close similarity of ὦ φίλ᾽ and ὠφελ- led, as
usual, to the omission of the one. When ὠφελῶν
alone was read, the line was easily seen to be metri-
cally defective, so some copyist who thought more of
sound than sense contrived to get good metre by the
easy expedient of lengthening ὠφελῶν to ὠφελούμενος.
Such a process of change is easy to parallel (e.g. the
similar but still more intricate one traced by Prof.
Jebb in Soph. O. 7. 1099—1100). The ὦ φίλ᾽ is
specially appropriate, for Neoptolemus has been
coaxing Philoktetes to come to Troy along with him,
his proved friend, φίλου wer’ ἀνδρός, as he says a few
lines above. And in his next words he assures
Philoktetes that he is his friend (σοί που φίλος γ᾽ &v),
and further on addresses him as ὦ τάν.
N. MAcnicon.
* *
*
PLATO, Rep. VI. 488ε, ὅπως δὲ κυβερνήσει k.T.A.—
The sense is, that the sailors (οἰόμενοι, asin the MSS.,
may be ungrammatical but all the same natural) do not
think that their star-gazing pilot will have either the
knowledge or experience how to steer the ship (μήτε
τέχνην μήτε μελέτην τούτου, ὅπως κυβερνήσει) Without
regard to their wishes. Stallbaum thinks we have here
implied the modern contrast between scientific know-
ledge and practical skill ; but this is entirely irrele-
vant and un-Platonic. It is not less un-Platonic,
though far more plausible, to suppose that Plato
asserts, in opposition to the disobedient sailors, the
fitness of the scientific pilot to join in the scramble
for power and to emerge from it victorious : and the
same criticism must be applied to the substantially
identical view of Grote and Mr. Sidgwick, though
they suppose that not the sailors, but the pilot holds
that his knowledge in no way fits him to overcome
the mutinous tendencies of the crew. The fact is
that the philosopher has left the region of common
sense altogether and is enforcing his paradox that
Power is Knowledge, and Knowledge is Power—power
even over the wills of our fellow-men ; he is insisting
that the mutinous sailors are wrong in supposing the
observer of the stars to be a mere dreamer unable to
handle the ships despite their opposition. The 1st
book of the Republic has shown us how Plato regards
the man of knowledge as a ruler (ἰατρός = σωμάτων
ἄρχων, κυβερνήτης = ναυτῶν ἄρχων), and a so-called
ruler as no ruler at all unless he be a man of knowledge,
since otherwise he defeats his own ends. The Polzti-
cus (to which Stallbaum refers but of which he makes
insufficient use) shows the same thing in the abstract,
viz: that Plato regards government as a part of
science, γνωστικῆ not πρακτική. Why he did so is
not hard to understand, though at first sight the
Aristotelian τὸ τέλος οὐ γνῶσις ἀλλὰ πρᾶξις seems NO
such recondite a truth. Perfect government, Plato
saw, like perfect living, manifestly depends on know-
ledge ; in fact it was the special function of Socrates
and Plato to bring out this dependence. Again—as
the Politicus shows—Plato’s analysis of life, less com-
plete than Aristotle’s, made ouly two divisions of it,
the Cognitive and the Productive which brings into
existence σώματα πρότερον οὐκ ὄντα, 258e. Lastly
ideal government and perfect knowledge tended to
become identified in Plato’s mind as the highest
things he knew and worthy of equal reverence ; he
could not, like Aristotle, have calmly put πολιτική
below θεωρία, though allowing of course the greater
attractiveness of the latter. The thought that
government was really the result of competing forces,
of opposing wills, was abhorrent to him. Here he
was less really wise than the disobedient sailors in his
parable, who thought there was but one way to
become master of a ship, namely to beg or intrigue
for it, and felt sure that the scientific pilot had not
learnt by τέχνη or μελέτη how to do that; and we
agree with them, seeing that, thanks to Aristotle, we
understand life better than Plato did and see that
knowledge is the least part of ruling, the essential
element being to make others obey you. When
Plato attempts to force experience to corroborate his
doctrine he has but poor success. For he can say but
this—that, assuming government to include the sup-
pression of opposing wills, the scientific pilot (it is to
be noticed he is not called ‘scientific’ but apxixds) or
the scientific doctor still truly governs, for in case
of need (ἐάν τις κάμνῃ, 489b.) the law of nature and
necessity (ἔχει φύσιν---ἀναγκαῖον εἶναι, 7b.) forces us to
come to him and obey him whether we like or not.
But some men have died rather than eall in a doctor ;
ships have foundered because the crews would not ac-
cept the direction of the one competent man among
them; and (to come to the sphere of government proper)
the Jonians failed in their revolt and lost their indepen-
dence because they would not be the obedient subjects
of Dionysius of Phocaea. The man of skill and know-
ledge must thrust himself to the front, and it is well
for the world that he sometimes does ; though often
elbowed out by noisy incompetence, it is happily not
true that a man with the stuff for command in him
(οὗ ἂν τῇ ἀληθείᾳ τι ὄφελος ἢ) never solicits power
(489c. ; ef. 347a).
J. SOLOMON.
* *
*
Herovotus II. 22,
κῶς ὧν δῆτα ῥέοι ἂν ἀπὸ χιόνος, ἀπὸ τῶν θερμοτάτων
[τόπων] ῥέων ἐς τὰ ψυχρότερα; τῶν τὰ πολλά ἐστι
THE CLASSICAL REVIEW. 49
ἀνδρί γε λογίζεσθαι τοιούτων πέρι οἵῳ τε ἐόντι, ὡς οὐδὲ
εἰκὸς ἀπὸ χιόνος μεν ῥέειν: πρῶτον μὲν καὶ μέγιστον
μαρτύριον, κ.τ.λ.
I have never seen and cannot devise any satisfactory
method of construing the second clause of this sen-
tence. As it stands, τῶν τὰ πολλὰ must refer back
to what precedes, but it is evident from the beginning
of the next sentence that some preparation for it is
wanted in this. I believe that τεκμήρια lies concealed
in τὰ and that τῶν therefore was not in the original,
as it is not in all MSS. ; a blot or gap in the MS.
might easily cause the corruption. But exactly how
the words between ψυχρότερα and πολλά ἐστι stood
it seems now impossible to say. Stein’s reading is
still more incomprehensible to me than the vulgate.
AS Pram:
* *
*
TRACES OF VERSE IN PRosE WRITERS. — Xen.
Cyropacdia iv. 6 ἃ δ.---κἀγὼ μὲν ὃ τάλας νεκρὸν ἀντὶ
νυμφίου ἐκομισάμην καὶ ἔθαψα τηλικοῦτος ὧν ἄρτι
γενειάσκοντα τὸν ἄριστον παῖδα τὸν ἀγαπητόν: 6 δὲ
κατακανὼν ὥσπερ ἐχθρὺν ἀπολέσας οὔτε μεταμελ-
ὄμενος πώποτε φανερὺς ἐγένετο οὔτε ἀντὶ τοῦ κακοῦ
ἔργου τιμῆς τινος ἠξίωσε τὸν κατὰ ys. The whole
passage from which the words are taken has ἃ strongly
pathetical and half-poetie ring (compare πολλὰ
γοωμένη in § 9), and the last few lines are quite in the
style of the sepulchral common-place. “Apri γενειάσκων
was a stock opening for an epitaph on a young man ;
see Kaibel, Epigrammata Graeca 345 : ἄρτι γενειάζοντα
k.T.A. Where he writes ‘Exodium carminis haud
rarum Theocriteum est xi. 9 ἄρτι γενειάσδων᾽ (ubi
vide Meineke); compare also Kaibel 100: ἄρτι
γενειάσκων K.T.A.; add. 5520 : [ἄρτι] γενειήσαντι k.7.A.
also Anthol. (Jacobs) ii. 262 : “Apts γενειάζων ὃ καλὸς
Kal στερρὸς ἐρασταῖς | παιδὸς ἐρᾷ Λάδων. σύντομος 7
Νέμεσις. In fact Xenophon’s sentence might be
turned into elegiac verse, almost without alteration as
follows :
“Apt. γενειάσκοντα γέρων τὸν ἄριστον ἔθαψα
παῖδα κομισσάμενος νυμφίου ἀντὶ νεκρόν"
ὡς ἐχθρὸν δ᾽ ὃ κανὼν ἀπολέσσας“ οὐκ ἀλεγίζει,
τιμῆς τὸν κατὰ γῆς οὔτινος ἀξιόων.
But I have not the least belief that Xenophon had
any one particular epitaph in his mind; his diction
in this passage is merely coloured by the reminiscence
of many a phrase belonging to the vocabulary of
mourning and of condolence. A similar account must
I think, be given of the subjoined passage also.
Plutarch ii. 463 A: διὸ τῶν μὲν ἀσώτων ταῖς οἰκίαις
προσιόντες, αὐλητριδος ἀκούομεν ἑωθινῆς, καὶ πηλόν, ὥς
τις εἶπεν, οἴνου, καὶ σπαράγματα στεφάνων, καὶ κραι-
παλῶντας ὁρῶμεν ἐπὶ θύραις ἀκολούθους. It would be
obvious to suspect this passage to be a paraphrase of
Comic fragment, if the words ὥς τις εἶπεν inserted to
explain πηλὸν οἴνου did not prove it to proceed from
the author himself. But evidently his mind is full
of the vocabulary and phrases which are expressive
of revelry, and it would be very easy to write the
sentence in verse as follows :
Αὐλητρίδος θ᾽ ἑωθινῆς ἀκούομεν,
καὶ πηλὸν οἴνου καὶ στεφάνων σπαράγματα
καὶ κραιπαλῶντας ἐπὶ θύραις ὑπηρέτας
ὁρῶμεν.
But the attempt is, I believe, simply misleading.
Ὁ ΡΟ; iis 225
Wordsworth’s
couplet :
εἰς ἴδιον ἐξέραμ᾽ ἐπιστρέψας κύων.
>
λελουμένη θ᾽ bs εἰς κύλισμα βορβόρου.
NO. XXX. VOL. IV.
suggested to the late Bishop
mind the remains of an Jambic
Ido not know whether his conjecture is rendered
more probable, or less so, from a circumstance which
I think I have observed in this Epistle, viz. that the
writer’s style and manner of diction seem coloured
by areminiscence of the manner and rhythm of tragic
iambic verse. Without difficulty one Sunday after-
noon, with very little transposition of words, I con-
trived to write a good many verses of 2 Peter in
clumsy iambic trimeters. Take these lines from 1,
19 foll. :
Ἔχομεν βεβαιότερον δὲ τὸν προφητικὸν
λόγον, ᾧ καλῶς ποιεῖτε προσέχοντες νόον
φαίνοντι, λύχνος ὥς, ἐν αὐχμηρῷ τύπῳ,
ἕως ἂν ὑμῖν ἡμέρα διαυγάσῃ
ἐν καρδίασι κἀνατείλῃ φωσφόρος“.
Or these from ii. 1 foll. :
Aad δ᾽ ἐγένοντο ψευδοπροφῆται τῷ πάλαι,
ὑμῖν δ᾽ ἔσονται ψευδοδιδάσκαλοί ποτε
οἱ καὶ παρεισάζουσιν αἱρέσεις κακὰς
τὸν ἀγοράσαντα δεσπότην ἀρνούμενοι,
ἐπάγοντες αὑτοῖς τὴν ἀπωλείαν ταχύ.
πολλοὶ δ᾽ ἀκολουθήσουσι τῶνδ᾽ ἀσωτίᾳ,
ὁδός τ᾽ ἀληθείας βεβλασφημήσεται.
πλαστοῖς γὰρ ὑμᾶς ἐμπορεύσονται λόγοις,
πλεονεξίᾳ σπεύδοντες, οἷς οὐκ ἔκπαλαι
ἀργεῖ τὸ κρῖμά γ᾽ οὔτε νυστάζει μόρος".
Or again these from i. 9 :
ὦ μὴ πάρεστι ταῦτα, κεῖνός ἐστι δὴ
τυφλὺς μυωπάζων τε καὶ λήθην λαβὼν
ἤδη καθαρμοῦ τῶν προτέρων ἁμαρτιῶν.
So many strange things have been said about the
style of 2 Peter, that I am not afraid of my suggestion
being thought too extravagant to be mentioned.
ΒΕ, L. Hicks.
* *
*
CrcERO, pro Sestio, 126.
‘Non modo gladiatores, sed ipsi equi gladiatorum
repentinis sibilis extimescebant.’ Surely the orator
must have written: ‘non modo equi gladiatorum,
sed ipsi gladiatores repentinis sibilis extimescebant.’
Under the circumstances mentioned the horses, not
their stolid riders, would have been the first to take
alarm. They must have been well-bred horses, and
it is perfectly certain that the well-bred horse of
ancient times was fully as flighty as the well-bred
horse of our own.
Joun BAXTER.
Mendham, New Jersey.
[But a circus-horse has pretty well got over his
‘flightiness,’ and would be slower than his rider to
understand the significance of a sibilatio. |
* *
Viner, Georg. iv. 511 foll. Ina paper read to the
Oxford Philological Society on the Birds of Virgil,
afterwards amplified into a chapter in A Year with
the Birds, 1 commented on these beautiful lines, but
without doing full justice to Virgil’s truthfulness.
Comparing the passage with Homer, Od. xix. 518 foll.
(where, in spite of the puzzling epithet xAwpyis, the
bird can hardly be other than the nightingale), I
pointed out that Homer has two touches which are
true to the life—viz. the variety of the nightingale’s
song, and its habit of singing in the recesses of thick
cover ; while Virgil has neither of these. I refrained
from suggesting, as I might have done, that in making
E
δ0 THE CLASSICAL REVIEW.
the bird lament the loss of its young populea sub
umbra, Virgil has apparently chosen the most thinly-
leaved of all trees, and the one which would afford
least cover and least shade, and thus seem least of
all to suit the well-known habits of the bird.
Since then I have learnt that the poet was after all
perfectly true to nature. While listening to a night-
ingale some time ago close to Godstow Bridge, near
Oxford, it suddenly dawned on me that this bird was
actually singing populea sub wnbra. The tall Lom-
bardy poplars, on a lower branch of one of which it
was perched, were growing out of a thick undergrowth
of willow and alder and other scrub in which beyond
doubt the nest was placed. What Virgil was thinking
of must have been surely a scene of the same kind—
thick scrub along the banks of the streams or ditches
of reedy Mantua, with poplars growing out of the
undergrowth ; the ploughman had discovered the nest
in the scrub, and the bird had retired to a poplar
branch to give voice to its grief, or possibly to cheer
its mate to fresh exertions. This passage, then,
which was perhaps the only one I found to suggest
the smallest suspicion that Virgil could write of
birds conventionally rather than truly, is in all prob-
ability a reminiscence of some actual experience of
the poet’s boyhood.
As I have mentioned the Homeric passage, I wish
to add that the word χλωρηΐς, as applied to the night-
ingale, is after all not so puzzling as it might seem.
It is true that the bird’s colouring is neither olive-
green nor pale ; the brown is of a reddish tint, and
in the tail during flight the red is obvious. But
when it is seen δενδρέων ἐν πετάλοισι καθεζομένη πυκ-
ἐνοῖσιν, 1.6. With the shadows of thick foliage falling
on it, the appropriateness of the epithet becomes
probable and even certain. I have constantly had
occasion to notice the difference between the appear-
ance of a bird ‘in the bush’ and of the same bird
‘in the hand’; and it is necessary to remember
(apart from the vagueness of Greek words for colour)
that the Greek poets, and even Aristotle himself for
the most part, described birds as they knew them
in their own haunts, and not as specimens in a
museum.
Whether this interpretation was exactly what was
meant by the scholiasts when they explained the
word by “ἐν χλωροῖς φαινομένη,᾽ and “ἐν χλωροῖς
διατρίβουσα,᾽ I will not undertake to say. Their ex-
planation is rejected by Buchholz (Hom. Real. i. 2,
123), on the ground that it does not suit with the
distinctness of Homeric epithets ; and he prefers to
seek (without much success) for another species whose
colouring would better suit the natural meaning of
the word. The last writer on the animals of classical
antiquity (Ὁ. Keller, Thicre des classischen Alter-
thums, Ὁ. 467) suggests the serin finch, a near
relation of the canary. No ornithologist, I think,
would admit the appropriateness of Homer's lines to
any finch: and I prefer to think that the poet was
thinking of the true nightingale, and that the
scholiasts were not far wrong.
W. WaARDE Fow Ler.
* *
*
HorAct.—d47s Poetica, 349.
Nam neque chorda sonum reddit quem vult manus et
mens,
Poscentique gravem persacpe remittit acutum ;
Nec semper feriet quodeumque minabitur arcus.
There are three objections to the middle line of
these three:
(a) Horace is speaking of small faults, ‘quibus
ignovisse velimus.’ But this line describes a very
serious one and is therefore inappropriate.
(8) If this line be kept, either Horace was egre-
giously ignorant of the musical art or his contempo-
raries were enormous bunglers at it. What should
we think of a poet nowadays who should assert that
a violinist (any violinist) often played by mistake
on the G string for the A? Yet this is just about the
same thing. Ἶ ἢ
(γ) The omission of semper in the first line is quite
intolerable if we have to wait for another whole line
before it comes ὄψε κακῶς in the last. But if the
middle line is omitted, the semper easily goes with
both reddit and ferict.
There are strong reasons then for regarding line
349 as an interpolation inserted to explain the
tolerably perspicuous line which precedes.
Ἂς PrAiT
* *
ΕἼ
Propertivs II. xxxiv. Lines 30, 34, 40, contain
a strange repetition which gives evidences either of
corruption, or of the unfinished and unsettled state
of the poem.
30 has magno — amore.
34 has magno — amore in exactly the same position
in the pentameter.
40 has magno also in same place in the line.
43. includere. P’s componere evidently comes
from line 41, to which the scribe’s eye was attracted
by the similarity of the endings thwrno and torno.
70. Paley’s note is: impressis = non pressis,
νημέλκτοις. Can this be correct ? Surely impressis =
simply pressed.
91. Lycoride—vulnera is what Postgate calls a
sub-construction bracketed off from the rest of the
sentence, = vulnera a Lycoride data. The participle
is omitted as in IJ. xxxii. 23, rumor ad aures
(adlatus). 5. E. WINBOLT.
os
‘TrRisTIA.—Mr. Owen in his recent edition has
suggested that ‘ tristia’ is a translation of μέλεα and
that μέλεα should be read for μέλη in the list of Cal-
limachus’ works given by Suidas. There are at least
three objections to this view :—
(a) μέλεα is a poetical word and the titles of books
among the Greeks were rarely, if ever, fanciful or
sensational. ἀπολοφυρμοί or the like is what one
would expect.
(0) μέλεα would be ambiguous and a poet who chose
it for a title would invite his enemies to say that his
poems were κάρτ᾽ émévvua—miserable, infelicitous
productions.
(c) There is no great difficulty in explaining Μέλη
in Suidas. See Schneider, Callimachea, II. p. 18.
But though Callimachus never wrote a poem called
“ μέλεα,᾽ it is still possible that Ovid was translating
μέλεα. There seems to have been a belief, in Lesbos
if not elsewhere, that μέλη was to be derived from
μέλεα. The authority for this is a Lesbian mytho-
grapher Myrsilos, in Cramer’s Anecd. Oxon. I. p. 285 :
Μυρσίλος δὲ Tas ἐν Λέσβῳ γενομένας παρθένους Μούσας
ἐπὶ τὰ πένθη φοιτᾶν καὶ θρηνεῖν: ὕθεν ἐπεκράτησε τὸ τὰ
ἀδόμενα μέλεα κληθῆναι. If Ovid knew this story, it
may have suggested to him his title ‘tristia,’ or he
may have known the etymelogy and not the story ;
for if he had known the latter, he would almost cer-
tainly have alluded to it (the Muses were trained by
Megaclo to appease by their singing the wrath of
her father Macar: see last number of the Philologus
where the Lesbian story of the Muses is discussed at
length and from which I take the reference for Myr-
silos). But all this is baseless conjecture.
W. ἢ, Harpie.
* *
*
THE CLASSICAL REVIEW. 51
LIDDELL AND ScorT’s GREEK LEXICON : Corrigenda
ΤΙ.
ἀγκύλη. 2.—Under this heading we find the phrase
ἀγκυλόγλωσσον πάθος, explained as ‘a similar disease
of the tongue’: but ἀγκυλόγλωσσος deserves a
separate place, such as it has in the Dictionaries of
Hedericus, Alexandre, &c. In Hed. we find ἀγκυλό-
yAwooor, ov, τὸ, linguae impedimentum: ἀγκυλό-
yAwooos, ov, ὃ καὶ 7, pracpeditam habens linguam.
Alex, gives first the adj., which he explains ‘affligé
dune ankyloglosse ; ii. subst. ankyloglosse, vice de la
langue génée par le filet.’
ἀκηδής 11.---11. 21, 123 is cited thus—oaw’ ἀπολιχ-
μήσονται : it should be αἷμ᾽ aoa.
ἁλίσκομαι 2.—In quotation from Xen. Hell. 1, 1,
23 the order of the words is inaccurate: it should be
γράμματα πεμφθέντα ἑάλωσαν εἰς ᾿Αθήνας.
auepyw.—In the quotation from Theocr. 26, 3 read
ἀμερξάμεναι, instead of masculine forra.
oud0ev.—The words cited from Od. 1, 10 are thus
rendered—‘ of which from what source soever...tell us
also.’ It should rather be (‘starting) from some part
(or ‘ point’) in these events’ &c. It was thus that
the Sch. understood 1{---ἀπό τινος μέρους ἀρξαμένη.
ἀπόζω.--- ον the construction of this verb with the
gen. reference should have been made to Luc. Dial.
Mf. 1, 5.
apreuns.—In the quotation from Od. 13, 43 the
order of words is incorrect; read σὺν ἀρτεμέεσσι
φίλοισιν.
βασανίζω II. 3.—The metaphorical meaning of the
verb is confined in L. and S. to the perf. pass. part. ;
but reference should be given to Longinus de Subl.
10 sub fin. τὸ ἔπος ὁμοίως ἐβασάνισεν.
BiBnut.—The form βίβαντι is represented as 3 plur.
on authority of Epigr. Lacon. in Ahrens ; but Ahr.
11, 312 describes it as ‘tertia singularis,’ which is
evident from the epigram (see Pollux 4, 102)— xiar
ἅδε ποκὰ βίβαντι, πλεῖστα δὴ τῶν πήποκα. The same
error is committed by Veitch, Greek Verbs, p. 116,
though probably one was copied from the other.
Taijios.—In quotation from Od. 7, 324 we find
Taniov υἱόν (where the dizresis is marked, thongh not
at beginning) ; read Γαιήιον.
δάμνημι.---Ἴ1. 14, 199 is misquoted in the word
ἀθανάτως : it should be ἀθανάτους.
5wuds.—The authority quoted is ‘ Epilye. Kor. 2’ ;
but the title of Epilycus’s play was Κωραλίσκος : the
abbreviation therefore should be Kwp. It will be
found correctly given s.v. Bapaé.
ἐνίημι 11. intr.—After the citation from Xen. Cyr.
add ib. Hell. 2, 4, 32.
ἔρρω II. 3.—In the quotation from Xen. Hell. 1,
1, 23 ‘ Mindarus’ is credited with the famous laconic
despatch which announces his own death. For ‘ Min-
darus’ read ‘ Hippocrates.’
érepdyAwooos.—In the quotation from 1 Ep. Cor.
xiv. 21 we read ἑτεγλώσσοις : it should be ἑτερογλώσ-
cools.
khp.—For the phrase περὶ κῆρι φιλεῖν, 71, 4, 46 is
referred to ; but Hom. there has περὶ κῆρι τιέσκετο.
Here also L. and S. make πέρι parox., whereas under
περί E they give the same phrase with oxytone.
Uniformity at least should be observed in the accen-
tuation. In 71. 13, 480 περὶ κῆρι φίλησε is found.
κιγκλίς I1.—For sense Ξ- δρύφακτοι we are referred to
‘Id. Caes. 68’; but the Jast reference was C.J. 481.
Read Plut. Caes. 68.
veoxuow.—In quotation from Thue. 1, 12 we read
νεόχμωσε, the augment being omitted ; it should be
ἐνεόχμωσε.
οἰκεῖος B.—oixelws συνεῖναι is given as the equiva-
lent for Lat. familiariter uti: but a closer reference
is Xen. Hell. 2, 3, 16 οἰκείως ἐχρῆτο τῷ Θηραμένει.
παντελής I1].—For παντελῶς with adjs. we are only
referred to fragments of comic poets; but cf. Xen.
Hell. 2, 4, 34 π. βαθεῖαν.
περί E.—rrept κῆρι φιλεῖν is supported by 17]. 19,
119; but there Hom. writes νεμεσσῶμαι περὶ κῆρι:
the reference should be 71. 13, 430.
oipds.—Reference should be given to N. T. 2 Pet.
2, 4, where the best MSS. read σειροῖς 1.6. ‘dens,’
instead of Vulg. σειραῖς. So Alford and Tregelles,
while Lachmann and Tischendorf read σιροῖς.
Reference should also be made to Dem. 135, 5 and
for the form σειρός (a v.d. in Dem.) Pollux 9, 49 ;
Phot. 504, 23; Varro R. R. 1, 57 and 63.
ὕω 2.—*So the Greeks used vege’ : read viper.
φιλέω.---Οα. 14, 146 is cited thus ὃν πέρι κηρὶ φ.
Ζεύς : but Hom. here has πέρι yap μ᾽ ἐφίλει. More-
over the accent is incorrect : it should be κῆρι.
Launcetor 1). DowDALL.
CLASSICAL EDUCATION IN THE UNITED STATES.
Ill.
In my two former letters (pp. 77-80, 223-225) a
description was attempted of the general conditions
under which classical studies are pursued among us
and of the work and aims of our representative
schools. It now remains for me to speak in this, the
final letter of the series, of classical studies as pursued
at college and university, 1.6. of the general training
in these subjects received by candidates for the B.A.
degree, and of the special training received by candi-
dates for the higher or professional degree of Ph. D.
This account will be followed by a few remarks on
the characteristics of our classical scholarship and on
our leading philological associations and publications.
The American youth who comes up to college to
receive a classical training has seldom carried his
work at school in Greek and Latin beyond the mere
requisitions for admission to college: he knows
something of Greek and Latin grammar; he will have
read a few books of Xenophon’s Anabasis, two or three
of Homer's Z/iad, and perhaps a little Herodotus; two
or three books of Caesar, several orations of Cicero,
five or six books of Virgil. If he comes from our
better schools he will have written a little English
into Greek and English into Latin—continuous nar-
rative in preference to disjointed sentences—enough
to show that he has a practical familiarity with
forms and the leading rules of syntax. He will also
know something about Greek and Roman history.
At a few of the better colleges he will be required to
show his ability to translate ‘at sight,’ without the
help of grammar or dictionary, passages of average
difficulty from Attic prose (usually taken from
Xenophon), and from Homer, and also from Latin
prose (Caesar) and from Virgil: in these cases the
quantitative requisition of books is usually waived,
ability to read independently being alone insisted
upon. At the leading colleges (Harvard, Yale, ete.),
a strict examination on these subjects is prescribed
E 2
52 THE CLASSICAL REVIEW.
for candidates ; many of the minor institutions, how-
ever, as was pointed out in my first letter, admit
students without examination to the Freshman class,
on the certificate of their teachers at school that they
are qualified to enter college. That the quality of
the preparation for college, as also of the work done at
the college, differs very widely in different parts of
the country goes without saying: in the absence of
common supervision each school and each college is
a law unto itself. The attainments in classical scholar-
ship of graduates of equal ability and industry from
different colleges differ vastly, and this difference
makes the problem of university or ‘graduate’ instruc-
tion a difficult one. [Ὁ not unfrequently happens
that the graduates of one college are found hardly
more than ready to begin the classical studies in
other colleges. ‘This unevenness of classical work
throughout the country is a feature to be deplored ;
there are, however, encouraging indications of im-
provement on all sides, which lead friends of classical
scholarship to believe that our standards of education
are improved and are everywhere continuing to im-
prove. The most casual comparison of the curri-
culum in classical studies of our colleges thirty or
forty years ago with that of to-day will convince
one of this fact. It will be impossible in this letter
to take note of these differences: we must confine
our remarks to a desciiption of the work and aims of
the better colleges.
In our colleges (see p. 78) the course of study for
the B.A. degree extends through four years, the
first two years (Freshman and Sophomore) being
mainly given—where there is a fixed curriculum of
studies—to the classics, mathematics, and modern
languages, and the last two years (Junior and Senior)
to the study of certain of the physical and natural
sciences, to philosophy, literature, history (political
and literary), political economy, ete. This pro-
gramme is greatly modified in colleges where freedom
in choice of studies either entire or with limitations
is granted the students. Except with students
who wish to make something of a speciality of
the classies, Greek and Latin are not carried beyond
the close of Sophomore. In fact this period marks
off a distinct stage in college life: in institutions
which are able to provide a sufficient number of
special courses, it is recognized as the point at
which a student may with profit begin to specialize
in his work—earlier specialization, at least in the
majority of cases, being regarded as premature. The
average student’s attainments at the end of his
Sophomore year bear some resemblance to those of
a German youth at the close of his course at the
Gymnasium.
In the Freshman year school-methods of instruc-
tion prevail. Freshman classes are divided into
small sections (of not more than 25 or 35) which
receive from three to four hours of weekly instruction
in Greek and Latin; the same amount of time is also
given to these subjects in the Sophomore year.
Lessons are regularly assigned and ‘recited,’ 7.e.
there is-a daily oral examination on the work set for
the day, with vivé voce translation. The instruction
given by the teacher is usually in the form of run-
ning comment upon the passage read, with an occa-
sional more formal lecture. The same methods pre-
vail to a certain extent in Sophomore year, though
here the range of instruction widens: the work of
the school and of the Freshman year having developed
in the student a command of the language, new
points of view, now made attainable, are taken ; the
individual form or style of the authors read, their
subject-matter, and vistas of ancient life and thought
thereby opened up, are made fruitful and stimulating
subjects of instruction.
In the Freshman year, which is not wholly unlike
Ober-Secunda or Unter-Prima in Germany—except
that our Freshman will average 191 years of age—
the usual plan is to divide the line somewhat as
follows :—in Greek between Homer and Attic prose
(history or oratory), and in Latin between Livy,
Cicero and Horace’s Odes. In the Sophomore year
two or three Greek dramas are read, and certain
speeches: in Latin, Horace’s Satires and Epistles,
two or three plays of Plautus or Terence and some-
thing from Tacitus or Pliny. The programme is
greatly varied in actual practice.?
Besides the reading of the prescribed books, and
gaining a general knowledge of the topics suggested
by the reading, the average classical student by the
end of his Sophomore year has done but little. In
the composition of Greek and Latin prose he has
advanced only slightly beyond the school stage. At
colleges where Honours in Classics are arranged for,
and in the examinations of which there are papers in
Greek and Latin composition, he will have done
more: here he will be able to put historical English
prose into passable Greek or Latin, with no very
glaring faults of style. At no college—and of course
not at school—will he have learned verse composition:
in fact Greek and Latin composition has never formed
any essential part in our programme of classieal
studies, though here and there verturesome teachers
have offered instruction in it. Whether our classi-
cal scholarship has gained or lost by this omission
cannot here be argued ; that, however, from its
failure adequately to develope facility in prose com-
position, it has lost alike in accuracy and depth, no
one can dispute.
As was remarked above, with the close of the
second of their four years in college the majority of
American students of the present generation have
done with their classics. Those only continue (in
such colleges as provide for advanced work) in whom
a love for the subject has been awakened ; many of
these intend to become classical teachers. This change
in the number and character of students radically
affects—or should affect—the method of instruction.
Progress is made on the lines struck out in the Sopho-
more year toward a more complete independence in
the student’s activity : the student’s attitude towards
his books and his teacher gradually changes from
that of a school-boy learning lessons under a master
to that of an independent worker engaged in a com-
mon enterprise with an older,more experienced friend
and guide. The books and subjects with which clas-
sical students occupy themselves during these last
two years in college vary greatly, as also the method
and aims of classical teachers. In that large number
of colleges where the provision for advanced classical
1 At Yale College in 1888-89 the programme for
Greek and Latin for the Freshman and Sophomore
class was as follows :—
FresHMAN YEAR. Greck.—Homer’s Iliad (two
books) ; Homer’s Odyssey (four books) ; Xenophon’s
Hellenica (50 pp.) ; Herodotus (50 pp.) ; Lysias (80
pp.) ; Prose Composition. Latin.—Cicero’s Tusculan
Disputations (book i) and Selected Letters; Livy
(books xxi, xxii); Satires of Horace: Prose Compo-
sition ; History of the Roman Republic.
SopHomorE YEAR. Greek.—The Antigone of
Sophocles ; the Agamemnon of Aeschylus ; the Medea
of Euripides ; the Panegyric of Isocrates ; Thucydides
(45 pp.); Plato’s Apology and Crito (the Frogs of
Aristophanes witha special section); Latin.— Agricola
of Tacitus; Selections from Pliny’s Letters; Odes
and Epodes of Horace; Trinwmmus and Miles
Gloriosus of Plautus ; sight reading from Cicero and
Sallust.
THE CLASSICAL REVIEW. 53
instruction is necessarily limited and where only a
few courses are offered, these courses will be confined
to the leading authors and subjects. In Greek the
serious study of Demosthenes and of Attic oratory,
of Plato and the history of Greek Philosophy—Aris-
totle is almost never read in American colleges—and
of a number of the more difficult Greek dramas will
be undertaken. In Latin the historians, Juvenal,
the philosophical works of Cicero, the more difficult
comedies of Plautus and Terence, together with liter-
ary history, form the principal subjects.
In the larger colleges where a considerable number
of elective courses in classics are offered, the variety
in subjects is much greater, as also the degree to
which specialization is carried. To the books and
subjects named above will be added eritical work in
various other branches of classical literature and
philology, as well as in comparative philology in its
relations to Greek and Latin. The more enterprising
of our classical teachers, many of whom are scholars
who in their professional training either in Germany
or in this country have become interested in some
author or subject, frequently take occasion to carry
on original work in their specialities by offering
courses in them in which they are aided in their
researches by their better pupils.?
The attainments of the students who have con-
tinued their classical studies up to the time of taking
their first degree in arts will necessarily vary greatly
in the different colleges and in the same college. In
the better institutions such a student will have had
the opportunity not only of reading widely and under
good guidance, but also of becoming something of a
specialist in a particular field of classical studies.
If we may continue the comparison made above, the
best of our men, at graduation from college, on an
average stand on a level with the young classical
philologian in the German university a year or two
before he takes his doctors degree; our B.A.’s
frequently earn this degree upon the close of two or
three years of advanced study in Germany. If the
American student pursues his studies at a college
where Honours in Classics are provided for, or where
scholarships, fellowships, or other prizes are open
to properly qualified candidates, and has duly
received those several recognitions, his work and his
attainments can be more definitely described. Only
a few of the leading colleges encourage classical
studies by provisions of this sort.”
1 As a rule American college teachers have no large
amount of leisure for private work. From twelve to
fifteen hours a week are spent on the average in the
lecture or ‘recitation’ room : in some colleges twenty
hours of class-room work in the week is not regarded
as excessive,
The number of classical teachers in the different
colleges varies with the number of students: in the
smaller colleges two ‘professors’ will give all the
instruction in Greek and Latin ; ordinarily, however,
there are at least two instructors, a professor and a
tutor, for each of these departments, and in the larger
colleges the classical staff will be made up of as many
as ten or a dozen officers. It is not usual for an
instructor to teach both Greek and Latin.
2 At Harvard there are two grades of Honours in
Classics, second-year Honours and final Honours.
The former are open to candidates not earlier than at
the close of their Sophomore year, and are awarded
on two conditions. The first condition is distin-
guished exceHence in classical studies amounting
to four courses ; the second condition consists in pass-
ing with distinction a special examination directed
to testing (a) the candidate’s ability to translate
The proportion of students reading for these Hon-
ours to the whole number of candidates for the B.A.
degree is extremely small. At Harvard, for example,
out of a Sophomore class of 251 only 14 obtained
second-year Honours ; and of a graduating class of
237 only 4 obtained final Honours. Scholarships
and fellowships are awarded on very different condi-
tions in the several colleges: the former carry with
them a smaller stipend (from $50-$350 annually),
and are given to undergraduates on evidence of merit
shown either at the regular examinations in the stipu-
lated departments or at special examinations. Fellow-
ships bring a larger stipend, ranging from $400 to
$1,000 per annum, and are commonly awarded for
not more than one year (renewable once or twice) to
very recent graduates of marked attainments or
promise, and frequently without special formal ex-
aminations. The Fellows are expected to pursue the
studies of their department either in residence at
their college or at some European university, From
approved Fellows the colleges usually recruit their staff
of instructors. It is perhaps needless to add that
these ‘Fellows’ stand in a relation to the corporate or
governing body of their college which is wholly dif-
ferent from that of the Fellows of an English college
to their Alma Mater.
Having taken note of the characteristic features of
classical education at our colleges, it now remains to
add a few remarks on the provisions made in the
United States for the advanced instruction of college
graduates, an instruction that might be termed pro-
fessional, inasmuch as it is given mainly to those who
have in view the profession of teaching.
The body of teachers in our colleges and secondary
schools is recruited in a variety of ways. ‘The
secondary schools are commonly satisfied if they
obtain as classical instructors promising college
graduates, who may have received no training beyond
that received by Bachelors of Arts. The colleges also,
until within a recent period, have asked for nothing
more ; at the present time, however, among candi-
dates for positions in the more reputable colleges, one
who has received a special training, such as a Doctor
of Philosophy, has, other things being equal, a dis-
tinct advantage.
The ‘graduate,’ (sometimes even called ‘post-gradu-
ate’) work of our candidates for the Ph. D. degree
is carried on either in Europe or in the United
States. The number of our graduates who continue
their studies in German universities is very large, a
number sufficient in any one year to fill a university
of very respectable size. The life, with its studies
and training, of the American student of classical
philology in Germany does not differ in any respect
Greek and Latin at sight (passages taken only from
the leading classical authors) : (b) his ability to write
Greek and Latin prose: (c) his knowledge of Greek
and Latin Grammar. Final Honours in Classics
awarded at graduation are given for great proficiency
in Greek and Latin ; the range of reading is extended
and a much more particular acquaintance with all
the related topics is expected: a Latin thesis, giving
evidence of original research, is also required.
3 Asarule the M.A. in the United States means
little except that its holder has at some time received
a B.A. degree. At the better colleges the candidate
is required to show that he has paid some attention
to liberal studies for a year or two after receiving his
B.A. degree. At the university of Virginia, however,
this degree has a singularly honourable position ; the
conditions on which it is bestowed are severe, and
it stands for distinct and high attainments in scholar-
ship.
54 THE CLASSICAL REVIEW.
from that of his German associates, which has been
well described by Professor Ziegler in the last volume
of this Review. While undoubtedly a certain amount
of indulgence is granted by university authorities to
all foreign students, it is still true that at the best
institutions the standards are not lowered for their
benefit. The fact that a large proportion of our
foremost classical scholars have received much, if
not the whole, of their university (7.c. post B.A.)
training in Germany has its distinct influence on the
character of our higher scholarship. The methods of
more elementary instruction at school and college are
perhaps not affected. But in the higher lines, in the
work of the university lecturer, in the favourite sub-
jects treated, and in the methods of treatment, the
influence of German university traditions is most
striking. The American has more or less of origin-
ality, but as a rule our advanced work in classical
philology is after all the German system adapted to
American conditions.
Provision for ‘ graduate ’ work in classical philology
is made by several of our colleges and universities,
both by setting apart a number of advanced and
technical courses, open only to approved B.A.’s,
mainly candidates for the higher degree, and by the
appointment of professors whose sole duty it shall be
to give graduate instruction. The universities where
provision of this sort has been made with greater
amplitude than elsewhere are Harvard, Yale, and
Johns Hopkins, to name them in the order of their
age. The graduate department in each of the two
former universities is the recent outgrowth of the
undergraduate department or college proper, while
that of the Johns Hopkins University is in the original
foundation. At Harvard and Yale the lines of separa-
tion between the higher undergraduate and the
graduate courses are not sharply drawn. In_ the
number of graduate courses offered and of officers
giving them, Harvard at the present time is some-
what in advance; but in point of numbers of students
pursuing graduate courses in classical philology, the
Johns Hopkins University has the advantage over all
other American institutions.
An outline of the work done at the latter univer-
sity, condensed from the official statements, with
supplementary remarks on graduate instruction else-
where will sufficiently bring out the main features of
our best professional training in classical philology.
The instruction in Greek at the Johns Hopkins
University is directed by Professor Gildersleeve. His
own courses are given in a seminary which consists
of the director, fellows and such advanced students
as shall satisfy the director of their fitness for an
active participation in the work by an essay, a critical
exercise, or some similar test of attainments and
capacity. All graduate students however, have the
privilege of attending the course. Each regular
member is required to take his turn as interpreter,
critic, analyst; and special fields of research are
assigned according to progress and bent. The student’s
private reading is so directed that in the usual period
of preparation for the higher degree a comprehensive
knowledge of Greek literature and Greek life may be
gained. By these arrangements the students are
brought into closer relations with the professor and
encouraged to perform more independent work and
engage in more extended experiments than would be
possible on a system of mere recitation or the simple
hearing of lectures. Each year some author, or a
group of closely connected authors, is made the centre
of work : thus in 1878-79 Lucian ; in 1884-85 the
Attic Orators ; in 1885-86 Plato and the literary
form of Greek philosophy ; in 1886-87 Aristophanes ;
in 1887-88 the Greek Historians, etc. In addition
to the work of the seminary, courses of lectures are
given by the director and by other qualified persons
on a variety of topics. In Latin instruction is given
on a plan similar to that followed in the Greek
courses, the director of the seminary being Dr
Minton Warren.
In all this advanced work the importance of the
two aspects of classical studies, the scientific and
the literary, are kept constantly in view. ‘ Without
scientific study the cultivation of the literary sense is
apt to degenerate into finical aestheticism: kept
apart from the large and liberal appreciation of
antique life in all its aspects, the scientific study of
the classic languages divorces itself from sympathy
with tradition and relinquishes its surest hold on the
world of culture, on which the structure of the
university must rest... . . The principle of scientific
specialization, not only for the mastery of methods
but for the joy and profit of individual achievement,
and the other principle of wide and free vision must
abide.’
The Seminary, or Seminariwm, patterned after the
best German models, notably that of Friedrich
Ritschl, is a distinctive feature of the graduate in-
struction at our best universities, and the methods of
seminary instruction are adopted even in the later
stages of the undergraduate course. At Harvard the
Greck and Latin seminary is at present conducted by
Professors F. D. Allen and ἃ. M. Lane. Besides the
exercises of the seminary proper which in 1888-9 were
given to the text-criticism and interpretation of the
Alcestis of Euripides, and of Persius, graduate
students within their term of residence receive in-
struction from these gentlemen, from Professor Good-
win, and from other members of the classical staff in
a number of courses, among which may be named
courses on Aristophanes and Greek Private Life,
Aristotle’s Politics; Aristotle’s Poetics, Pindar,
Aeschylus ; Thucydides’ Speeches ; Critical Study of
Homer’s Iliad; Study in Greek History, from the
Sources ; Greek Epigraphy ; History of Greek Litera-
ture ; Legal and Political Antiquities of Greece ;
Quintilian, Gellius, Latin Inscriptions; Private
Life of the Romans ; Roman Religion and Worship ;
Philosophy among the Romans (Cicero and Lucretius);
History of Latin Literature ; Latin Grammar ; Greek
and Latin Prose Composition ; Comparative Philology
as applied to Greek and Latin ; the Elements of
Sanskrit, ete.
The degree of Ph.D. in course is the token of the
succesful completion of an approved course of graduate
study. The conditions on which it is bestowed in
the United States are not uniform. In general two
or more yearsiof graduate study are prescribed, one of
which at least must be spent in residence. The
plan of proposed studies must previously have
received the sanction of the proper authorities, and
in most instances must provide for three large, and
not necessarily closely related subjects; one of these is
the candidate’s chief subject; in which he must show
satisfactory evidence before receiving his degree that
he has carried on original research, the results of
which are presented in a thesis or dissertation, written
either in Latin, as at Harvard, or in English. Writ-
ten and oral examinations cover the entire programme
of work.
There is no uniform usage as to printing of doctor-
dissertations: a few of the more meritorious, not only
of the Johns Hopkins University but also of other
institutions, have received the hospitality of the
American Journal of Philology. Several universities
have lately undertaken the collective publication of
such contributions to classical philology as might be
made by their officers and students ; these University
Studies—as at Harvard, Cornell, and elsewhere—will
doubtless contain many doctors’ dissertations.
THE CLASSICAL REVIEW. DD
It is of course upon the instruction imparted at
university, college, and school, that our classical scholar-
ship is based, and itis mainly in intimate connexion
with these institutions that our classical scholars live
and, work. ‘There are, however, other agencies that
promote philological activity among us, such as the
associations maintained by our scholars and the philo-
logical publications issued by these associations or as
private ventures, which we should not leave without
mention in this survey.
The American Philological Association is a society
founded ‘for the advancement and diffusion of philo-
legical knowledge,’ and of the three hundred and
_ fitty members the majority are classical teachers. It
was established in 1869 and holds an annual meeting,
at a different place each year—usually at a college
town : a goodly number of papers are presented at
each session, on a great variety of subjects relating
not alone to classical philology in the broadest sense
of this term but also to comparative philology, to
linguistics and to the modern languages. Many of
the more important of these papers are printed in
full in the annual Z'ransactions, and abstracts of all
the papers appear in the Proceedings. Nearly all the
foremost classical scholars in the United States,
with a number from Canada, are members of the
Association and have contributed to its publications.
A large number of our younger men have made their
début at the meetings of ‘the Association in original
contributions to classical philology. The value of
this society in encouraging and recognizing indepen-
dent activity, and in bringing into personal acquaint-
ance very many persons interested in classical edu-
cation and in classical research cannot be over-
estimated.
The schoolmasters and other teachers in various
parts of our country are organizcd into State and
into National Teachers’ Associations. At the meet-
ings of these bodies, note is taken of many classical
matters, but usually this is done from the point of
view of pedagogy. Among the most useful and
influential of these organizations should be named
the Association of Colleges and Preparatory Schools in
New England, at the semi-annual meeting of which the
improvement of our classical education is a favourite
theme.
The union of our leading colleges in support of the
American School of Classical Studies at Athens—
which was founded in 1882, and has been since
maintained by the annual subscriptions of a dozen or
more colleges—is not without its distinct influence in
promoting among us the spirit of co-operation and
common interest. It is perhaps too early to speak of
the influence that the school itself has exerted upon
American scholarship: it is certain, however, that
the growing interest in classical archaeology at many
of our colleges is due to the school. The project of
founding at Rome an institute of classical studies for
students of Roman history and archaeology has not yet
been put into execution, though it has been seriously
entertained. The work of the Archaeological Institute
of America (founded in 1879), in promoting research in
classic lands, deserves mention (at Assos, Southern
Italy, etc.) ; much of the most important part of the
work done under the auspices of the Institute yet
awaits publication. The American Jowrnal of Archae-
ology (founded in 1885, and edited by Professor A.
L. Frothingham of Princeton College) is the official
organ of the Institute, as well as of the Athens
Scheol: this quarterly, in addition to printing many
of the more valuable papers of the Institute and of
the School in advance of their appearance in the
Papers, publishes a large number of original contribu-
tions to archaeology, classical and Christian, by
writers both at home and abroad.
Of philological journals in the United States the
number is very small. The educational journals
and the literary and theological reviews in which
classical subjects receive occasional notice are too
numerous to be mentioned here. Until about ten
years ago the periodical in which classical matters
received the fullest and most scholarly treatment was
the Bibliotheca Sacra (founded in 1844). In 1880
Professor Gildersleeve founded the American Journal
of Philology, and has remained its editor ever since ;
in the ten volumes of this periodical—with the
twenty volumes of the Transactions of the American
Philological Association—it is safe to say that we
have the completest record of the movements and
tendencies of our classical scholarship at the present
time, as well as an earnest of future progress and
achievement.
To characterize American classical scholarship in
a sentence or two is a more difficult task than to
describe the main features of American classical edu-
cation in numbers of pages. Until the second half,
or rather until the last quarter, of the present century
our scholarship can hardly be said to have had any
distinctive features. Before this time there were
eminent scholars who left their impress on teachers
and taught: men like Anthon, whose large but ill.
digested learning emptied itself into a vast number of
annotated editions of the classics, which in the day
of narrow reading and meagre libraries were not
without their use ; like Woolsey, who left the path
of brilliant promise as a Greek professor for that of
yet greater eminence as college president and as pub-
licist ; like Edward Robinson, who revolutionized, if
not created, the Science of Biblical Archaeology.
But these were sporadic cases, without their influence
on the general movements of thought. Within the
last twenty years, however, several distinct tendencies
have shown themselves in our scholarship, due in
large part to the personal influence of two or
three of our leading scholars as well as to the spirit
of the times. The most striking of these tenden-
cies is the bent for linguistic science, and for the
treatment of classical subjects from the point of
view of comparative grammar: it is without doubt
due, at least initially, to the influence of Professor
Whitney, who not only in his writings but as pro-
fessor in Yale College has impressed himself upon a
large number of our most thoughtful young men.
This tendency was reinforced by the personality of
Georg Curtius, whose lectures were largely attended
by American philological students in Germany.
A second feature, not wholly to be separated from
the one last named, is the fondness of our scholars
for all matters connected with the form of the Greek
and Latin languages, in other words with questions
of syntax. 1t is undoubtedly true that Professor
Goodwin’s Greek Moods and Tenses (first edition,
1860), an extremely popular book among our teachers,
and Professor Gildersleeve’s numerous articles (7 γαλι-
sactions of the American Philological Association, and
the American Journal of Philology) on Greek syntax,as
well as his activity as director of the Greek seminary
at Baltimore, have in great measure given rise and
direction to that general interest in these subjects
which entitles us to speak of it as a feature of our
scholarship. It is perhaps needless to add that the
influence of these scholars is by no means confined to
the field here mentioned ; to them and to others
who cannot here be named is greatly due the impulse
to a larger and more catholic conception of classical
philology of which we find many clear evidences at
every hand.
Classical scholarship with us hardly exists except
in connexion with education. The pursuit of it, as
an independent field of activity, with a view to
50 THE CLASSICAL REVIEW.
making distinct and positive contributions to science,
is as yet in its earliest stages. Hence the literature
of classical philology in the United States is mainly
that of college and school text-books (dictionaries,
grammars, annotated texts). Of these the number
and variety is great. There is a certain directness
and practical good sense in the American mind which
have hitherto shown themselves to best advantage in
the making of books—and alas! in the working up
of other men’s books—for use in school and college.
The defects of undue deference to authority—with its
attendant willingness to take nearly everything at
second hand and its disinclination to undertake
independent investigations and thus to gain indepen-
dent views—is unfortunately too often a character-
istic of our work. But there are signs of a better
day. Our ideals of classical education are being
raised and, what is of equal if not of greater import-
ance, our view of the meaning and scope of classical
hilology is becoming enlarged. We recognize the
imitations necessarily imposed by our remoteness
from European libraries and other collections—but
classical scholarship is more than palaeography and
epigraphy. The important books, alike the ancient
texts and their ancient commentators, and the monu-
mental modern works in which the great masters of
classical philology have illumined dark stretches of
ancient thought and activity, and disclosed unsus-
pected and most significant relations ; the leading
periodical publications, in which one may follow
modern scholarship in its swift advance and ever
broadening conquests—are all within reach of the
American philologian. If to the acute observation,
in their historical and formal aspects, of the phe-
nomena of language, and especially of grammar—that
θριγκὸς μαθημάτων, as Boeckh, quoted by Professor
Gildersleeve, has called it—which has given a deserved
preeminence to two or three of our scholars, we add
a broad, sympathetic and practical treatment of the
subject-matter of the ancient books, in its infinite
variety and range, in its relation to literature, literary
history, political history, art, philosophy and all
the other manifold manifestations of ancient thought
and action, there is no reason why American scholars
should not have an important part in the enlarge-
ment of the world’s common knowledge of antiquity,
and in making this knowlege efficient in enlarging
beautifying and rendering yet more significant this
our modern life and thought.
J. H. Wricur.
HARVARD UNIVERSITY,
November 21st, 1889.
C. G. COBET.
The following is the reply sent by the
University of Leyden to the Cambridge
letter of condolence on the death of Cobet,
printed in our last number.
Academiae Cantabrigiensis
Procancellario, Pracpositis, Professoribus, Doctoribus
Seed:
Curatores, Rector ct Senatus Acadenviae
Lugduno-Batavae.
Litterae ad nos missae, quibus dolorem declara-
vistis, quo ΟΟΒΕΤῚ Vos affecit obitus, haud leviter nos
commoverunt. Si quid est, quod animos nostros
afflictos erigere valeat, verba sunt ista humanitatis ac
benevolentiae plena, quibus maestitiam nostram ad
Vos quoque pertinere professi estis. Quod summi
Viri memoria, quem haec Academia amissum plorat,
in alma apud Vos Musarum sede pio gratoque animo
colitur, documento nobis est immortalia eius de lit-
teris antiquis merita toto pectore sentire Vos et
agnoscere.
Profecto luget Coperi e vita excessum quisquis
mortalium artes et litteras amat atque humanitatis
studia celebrat. Summa veneratione nomen eius
memorant omnes, qui recordantur quam multa studia
illa ingenio eius et industriae accepta referant.
Inter insignes laudes autem, quae ubicunque
terrarum singularis Viri memoriae tribuuntur, vix
ullae magis honorificae putandae sunt, quam quibus
Vos illam prosecuti estis. Cum enim litterarum apti-
quarum amore Britanni inter Europae populos
emineant, tum nobilissimam Vestram Academiam
praecipuis sedibus adnumerandam esse constat, ubi
earum studia alantur ac foveantur a Viris egregiis,
ingenii et doctrinae gloria praestantibus.
Britannos CoBeTus noster per totam vitam eximie
dilexit et admiratus est. Permulta eis se debere
agnovit. Bentleium, Porsonium, Dawesium magis-
tros suos appellare solebat. Nemo erat inter aequales,
cui tam studiorum communitate, quam amicitiae
necessitudine magis se coniunctum sentiret, quam
Carolo Badham.
Admirationi autem ac reverentiae, quibus Vestram
nationem CoBETus coluit, Vos, Viri Clarissimi, verbis
Vestris amorem et pietatem spirantibus dignas vices
rependistis.
Haec fidem nobis faciunt, summum Virum, cuius
gratissimam memoriam nos semper retenturi simus,
apud Vos quoque perpetuis posterorum laudibus
celebratum iri.
Cum autem sinceras Vobis gratias agamus quod
nostrum dolorem a Vobis haud alienum duxistis,
simul omnia laeta Vobis optamus et prospera.
Utinam per longam saeculorum seriem vetustam
suam gloriam tueatur illustrissima Academia Cantab-
rigiensis !
L. A. J. W. Suorer, Curatorwm Praeses.
A. P. N. Francuimont, Academiae Rector.
σι Κα. HorrMann, Senatui ab actis.
Datum Lugduni Batavorum
pridie Nonas Decembres
A. S. MDCCCLXXXIX.
THE CLASSICAL REVIEW. 57
OTT UA REY.
DR. POTTS.
By the death of Dr. Potts on the 15th of
November last, Fettes College has lost some-
thing more than its first Headmaster. The
man who successfully undertakes the
supreme care of any new school will
always have some of the honour which
belongs to a founder; and Dr. Potts had
the distinction of being entrusted with the
working out of what was, in Scotland, a
novel and, at first, a not very popular idea.
Dr. Potts was born in 1834 and was edu-
cated at Shrewsbury under Dr. Kennedy and
at St. John’s College, Cambridge. In 1858
he graduated a senior optime and second
classic, winning also the junior Chancellor’s
medal and becoming a Fellow of St. John’s.
After taking his degree he became a master
first at Charterhouse and then at Rugby.
From Rugby he came in 1870 to Fettes
College.
Although the school was the first Public
School of the English type in Scotland, it
can hardly be said that there was any active
demand for it. True it was that many
Scotch parents desired an English Public
School education for their sons. But this
desire was in most cases coupled with a
certain dread of provincialism which led them
still to prefer a southern school, especially
as to ordinary boys Fettes College offered no
advantages in the way of economy. It was
Dr. Potts’ principal task to win for the new
school a position distinct from that of an
English Public School inconveniently situ-
ated. He wished to acclimatize it as a
Scotch institution to flourish side by side
with the native public schools. In the most
jealous community to be found in the British
Isles this was an object which could not be
realized at once, and Dr. Potts never
claimed to have thoroughly realized it. But
he made great and striking progress towards
that end, and had the satisfaction long before
his death of witnessing the University suc-
cesses of many Scotch pupils who, had it not
been for Fettes College, would probably have
followed other paths.
Dr. Potts is perhaps best known to the
world as author of Hints towards Latin Prose
Composition, a book which was the first to
place the results attained by German erudition
in this field of research, by Nigelsbach and
others, within the reach of English students ;
and these results it presented in a fresh and
incisive form, re-casting them in the mould
which the fine literary instinct of English
scholarship can alone supply. Other books
on Latin prose style have since appeared,
each with merits of its own; but the Hints
may claim to have marked an epoch in the
teaching of Latin Prose such as no other
book has made. ΤῸ those who had been
reared on the old ‘ Latin Arnold’ it came as
a revelation. No one can read it without
gaining a clearer insight into Roman charac-
ter and its reflection in the language of Rome.
In his own Sixth Form teaching, how-
ever, Dr. Potts paid more attention to the
cultivation of Latin and Greek verse; with
what success the achievements in this line of
his old pupils at Cambridge will have shown.
In this no doubt he was true to Shrewsbury
traditions. But besides that, the natural
bent of his mind led him to set more store
by that literary sympathy with classical
feeling and that versatility of expression
which verse-composition especially brings
into play, than by flawless and terse diction.
As an educator he was brilliant rather than
methodical; he taught as he talked ; there
was an exuberant flow of fancy, wit, and
anecdote ; his mind was of that order which
is more quick to note resemblances than
differences ; he had a poet’s imagination and
an artist’s eye for beauty. Those who heard
him in school translate and almost declaim
his Demosthenes, translate and almost act
his Plautus and Aristophanes, could not fail
to catch some of his enthusiasm.
He was essentially conservative in his
methods and ideal of education. He taught
the classics as literature —always with a
view to knowledge οἵ men and life. Not
that he ever condoned inaccuracy or was
satisfied with the ‘ general sense’ in transla-
tion ; but he never forgot that a poem or
speech is a work of art which must be appre-
ciated as a whole, and not merely considered
in its details. Htymology was interesting
to him in so far as it illustrated the meaning
and history of a word, but he took no real
interest in the shifting phases of the science
of Comparative Philology, and refused to
treat the classical languages as a herbarium
in which dried specimens are kept in cases
and labelled for comparison with Erse or Old
Zend. Even in matters of grammar and
syntax he trusted more to his own fine in-
stinct for style than to methodical research
and tabulation of instances. Nor did he pay,
at any time of his life, any special attention
to the study of textual criticism. Some of
58 THE CLASSICAL REVIEW.
this may possibly be explained by want of
leisure. Fine scholar as he was, he was a
schoolmaster before everything else: he
threw himself heart and soul into the life of
the school, and this, as he conceived and
made it, was many-sided enough to engross
his energies.
He did not read a large number of books
with his Sixth Form, but what he did, he read
with minute care. He never ceased to insist
upon the necessity of private reading, and
frequently quoted in that behalf the precepts
and example of his own Headmaster. The
Sixth Form of Dr. Kennedy at Shrewsbury
was his ideal of what, for scholarly feeling,
a Sixth Form should be. He never wearied
of impressing upon his own pupils the indus-
try which had been required of himself
while a Sixth Form boy. For this purpose
he would recall, often with a very happy
representation of his voice and manner, how
after finishing a book of Thucydides Dr.
Kennedy would quietly say, ‘You can read
the other seven books in your studies.’
Apart from his power of teaching Dr.
Potts had great gifts as a Headmaster. He
was a man whose commendation could not
but be prized, and whose censure no one
could affect to receive without concern. No
man knew better than he what to pass over
and what to notice. Always open and
trustful, he was, even when finding fault,
courteous, considerate and moderate. Hence
his great and lasting power. There was no
reaction from his influence. His pupils
never outgrew his teaching. And he com-
manded respect by showing it.
No account of Dr. Potts, however brief,
would be complete without some reference
to his character as a preacher. The religious
situation of Fettes College rendered the
superintendence of the chapel service a
matter of the greatest delicacy. It was a
lay service designed to disarm the jealousy
of rival denominations. Nothing could
have been happier than the way in which
the exigences of the case were met both in
the pulpit and in the reading-desk. Dr.
Potts was a preacher of great power and
occasionally of true eloquence, and a reader
of more than usual dignity and impressive-
ness. The chapel at Fettes College became
popular with parents, and it is perhaps not
too much to say that it was there that much
of the prejudice which the school had to fight
against was removed.
To conclude this notice of a really gifted
teacher it only remains to say that his
interests and sympathies were universal.
As he was broad and general in his teach-
ing, so he was widely appreciative of merit.
He took pride in the success of his pupils in
whatever career it was attained. Good
work was to him good work on whatever
expended, He would mention the name of
a pupil who in a house of business had
gained the confidence and esteem of his
employers as proudly as that of a University
prizeman. And he was held in honour as
much by the one character as by the other.
BROWNING AS A CLASSICAL
No great poet has ever made himself
interesting to so many kinds of students, on
their purely technical and __ professional
ground, as Mr. Browning. Philosophers,
painters, musicians, Italian historians,
German historians, ecclesiastical historians,
Hebraists, all have a large part in him, and
certainly not the smallest share in the in-
heritance belongs to the student of Greek
(scarcely perhaps of Latin) antiquity. In
general, it may be observed, as an interest-
ing comment on the supposed ‘decline of
Greek,’ that the half-dozen great indisputable
poets of our time, all till yesterday writing
together, have been passionate Hellenists,
and not least popular when they were most
Hellenic, In this respect, if in no other,
Mr. Browning stood in a line with the
authors of Zithonus and Jason, Atalanta in
Calydon and the Strayed Reveller,
SCHOLAR.
In his writings there are two separate
strata of classical learning, sharply separ-
ated from each other—the original store of
memories that -he retained from his early
reading,{and the special studies of the period
that began with Balaustion’s Adventure
(1871) and ended with the translation of the
Agamemnon (1877) ; but the after-effects of
this second period re-appeared at intervals
for the rest of his life. In both these strata
antiquity is treated in a way peculiar to Mr.
Browning. No great poet ever absorbed so
much with so little effect on himself. His
classical learning, whether of his youth or
of his declining manhood, had no more power
than the Italian learning of his prime or the
Rabbinical learning of his old age to make
the least change in the ideals of his imagin-
ation or in his way of looking at life, nay,
not even in his literary form, whether of
THE CLASSICAL REVIEW. 59
composition or of style. Such as he came
forth full-grown in Paracelsus, such he re-
mained in Asolando, progressing royally
through innumerable worlds of imagination,
but in all of them not a colonist but a con-
queror, taking no impress from any, but
bending them all to the purposes of his
imperial will. In other words, we must go
to him, not to know what antiquity was like,
but what kind of thoughts it inspired in
him.
Of course the same thing is true of all
great imaginative artists ; their imaginative
re-constructions of the past are not to be
read as works of instruction for the sake of
information about the past. But Mr.
Browning seemed to.be bent on throwing
away the poet’s privilege, and presenting his
research to be judged on its own prosaic and
scientific merits. And so we are continually
distracted. The poet who stands the most
remote from the Greeks in his essential
character gives us the most of the Greeks in
outward detail. Any mind less compre-
hensive than his own must be content with
realizing alternately what he realized all at
once, on the one side the ‘then’ and ‘ there’
of Athenian civilization, drawn out into its
transitory and trivial particulars, on the
other side the essential spiritual issues,
which to us are not truly spiritual or essen-
tial until they are made freely and frankly
modern.
In the first period (1.6. before 1871) it is
easy to search out all his ancient references,
for they are all definitely,recognizable. He
adds a proper name or a description of an
incident ; it is the very rarest thing with
him to absorb an ancient phrase into the
current of his language without giving
notice of it, in the manner of Milton and
Tennyson. Attentive reading may discover
other instances ; the only one that I have
observed in this period is in Paracelsus (p.
83. ed. 1872), ‘Save a thin corpse at pleasure
of the wind,’ which I imagine to be a remi-
niscence of ventis Jludibrium datur and
αἰθέριον kivvypa. And the list of his refer-
ences 15 certainly equivalent to a statement
of the things which interested him in the
classics. Throughout his life, everything is
explicit with him ; he talks about everything
that he cares for. If we listened to his
earliest declaration in Pauline, we should
think that he loved all the Greek poets
alike ; there we have the ‘ high-crested chief,
Sailing with troops of friends to Tenedos,’
and ‘the king, Treading the purple calmly
to his death,’ and ‘how she, The fair pale
sister, went to her chill fgrave, With power
to love and to be loved and live,’ and the
hero even professes to have ‘lived with
Plato’ and to ‘have the key to life.’ But
in all the subsequent works, up to 1871, the
one ancient whom Mr. Browning really loves
is Aeschylus ; ‘the thunder-phrase of the
Athenian, grown Up out of memories of
Marathon,’ ‘ You, who, Plataea and Salamis
being scant, Put up with Aetna for a stimu-
lant,’ ‘the halt and maimed Iketides,’ ‘ who
made his Titan’s arch-device The giving men
blind hopes to spice The meal of life with,’
even St. John in the Death in the Desert
must refer to ‘that fable of Prometheus and
his theft,’ and yet further to ‘those satyrs
of his play, Who touched it in gay wonder at
the thing.’ There is nothing like the same
wealth of reference to any other author.
Homer is mentioned a few times with great
but rather conventional reverence (one
quotation in Pippa Passes); the future
destiny of Euripides is scarcely foreshadowed
in Artemis Prologizes (1842), and in the lovely
passage in Waring (1842) which summarizes
the history of Iphigeneia. Sophocles, as
we have seen, had his place in Pauline, but
he never appears again in all this period.
Plato, with whom Pauline’s lover had ‘ lived,’
onlyappears once: ‘ I would fain...in all God’s
acts (as Plato cries He doth) He should
geometrize’ (Christmas Eve and Easter-Day).
Besides direct reference to authors, there
are references to historical and mythical
story, but these generally look as if they
came from common knowledge rather than
from actual classical reading. One group of
such references is specially important, those
which rest not on literature at all but on
works of art. Even in Pauline the long
description of Andromeda, and the adjective
in the ‘naked Swift-footed,’ already show
that to the poet’s mind the ancient world is
embodied in statues not less than in poems.
After Pauline, his long years in Italy deep-
ened and deepened his pre-occupation with
art; all his conception of some important
Greek figures, Apollo, Herakles, Niobe, seems
to be derived from art, not from literature
at all.
It may be noticed also that not only an-
tiquity itself but the modern re-discovery of
antiquity had at one time a strong fascin-
ation for him. We cannot now recover the
amount of study which he had given to the
more northern and medieval side of the
Renaissance, because it was very soon ab-
sorbed by the overmastering attraction of
the same movement in Italy ; but it is em-
bodied beyond all possibility of under-estimate
in Paracelus and the Grammarian’s Funeral,
00 THE CLASSICAL REVIEW.
Paracelsus could only despise ‘ the meanest
plodder Trithemius quotes a marvel,’ but
that was not the judgment of Paracelsus’s
creator. Perhaps the philologist who mag-
nifies his vocation the most can scarcely take
to himself the lines that tell of one who
‘knew the signal, and stepped on with pride
Over men’s pity,’ who ‘throws himself on
God, and unperplexed Seeking shall find
Him,’ whose place was ‘ where meteors shoot,
clouds form, Lightnings are loosened, Stars
come and go.’ But there the lines remain, a
trumpet-call to the greatest of us and a
consolation to the meanest, the everlasting
charter of the scholar.
As we have seen, a new period begins in
1871, the period of minute study of Euripides
and Aristophanes. It might be suggested
that he had studied them before, but I must
repeat the remark that he would have men-
tioned them if he had. We must take his
new departure in writing as a new departure
in reading. It can be no secret how
Balaustion’s Adventure came to be written.
After Mrs. Browning’s death he never
shrinks from acknowledging her inspiring
presence in all his work, and in this case he
takes pains to make the connexion clear by
the prefixed motto. But in truth the motto
was not wanted. In the epilogue, where
Balaustion lets her fancy play round the
story of Alcestis, the poet is revealing the
story of his own life. Alcestis must be sent
back to the earth because when she left
Admetus the powers of her soul were added
to his, and he became strong beyond human
due. ‘Two souls in one were formidable
odds. Admetos must not be himself and
thou.’ Under that figure he has shown us
his own soul, passing through those twenty-
eight lonely years strengthened with the
strength of the Alcestis who did not
return.
Coming to the study of Euripides in this
way, as a sacred legacy from Mrs. Browning,
he could not fail to see in him higher things
than the mass of critics have seen. Where
Euripides gave a comic turn to the traditional
mythology, it was to bring the underlying
solemnity into sharper prominence ; where
he marshalled the opposing arguments with
wire-drawn volubility, it was to follow up
the self-deceiving soul into its remotest
hiding-place. Let us not say that this was
wrong. A poet who is also a philosopher
and a prophet has a right to see philosophy
and religion that we cannot see in the works
of another poet. In Mr. Browning’s case
the vision was helped by a theory which he
seems to have held with ever-strengthening
conviction, that a poet who could move him
to admiration must be fundamentally the
same kind of poet as himself. However
this may have been, he was inevitably led
from Euripides to his contemporary critic ;
and so he gave two or three years to that
exhaustive reading of Greek comedy, with
all its scholia, fragments, and commentaries,
which is turned out pell-mell in the pages of
Aristophanes’s Apology. Within its own
field, the poem is the most astounding monu-
ment of erudition. Who was Saperdion 4
Where in Greek literature does νεβλαρέται
occur as an exclamation? (Mr. Browning
once answered this question by referring a
correspondent to a fragment of Alexis,
before the word had made its way into
Liddell and Scott.) The erudition is astound-
ing, certainly. But is the main contention
tenable? Can we imagine Aristophanes de-
fending his artistic method as he does here ?
Nay, can we imagine him defending it at all?
Mr. Browning errs in good company, with
Heine and Mr. Symonds and Mr. Swinburne,
but the truth is that the Weltvernichtungsidee
is a mare’s nest. All these learned and
ingenious critics have gone astray because
they could not bring themselves to realize
that the ‘secret of Aristophanes’ is a secret
of pure and simple frivolity. To go back to
a former remark, Mr. Browning has drawn
out the spiritual issues which were involved
in the contest between comedy and _ philo-
sophic tragedy ; but as to understanding the
personality of Aristophanes, that would be
better done by Mr. Burnand.
It is right also, and surely not irreverent
even at this moment, to point out that Mr.
Browning’s Greek learning is as narrow as
it is wonderful. We might have thought
that Aristophanes and Euripides would have
led to some of their great contemporaries.
But πο. Aeschylus Mr. Browning knew
already. Sophocles of course has to be
brought in, but only with the indispensable
minimum of reference ; Plato is mentioned
half-a-dozen times (always as ‘ Aristullos ’)
but only in connexions which suggest books
on Aristophanes rather than his own writ-
ings; where Socrates is mentioned, no say-
ing recorded by Plato or Xenophon is ever
quoted; the one mention of Thucydides
(‘Thoukudides invent his epitaph’) is
obscurely worded, but I cannot help suspect-
ing that Mr. Browning had forgotten his
exile.
In short, Mr. Browning’s knowledge of
Greek literature was narrow. But that was
his way with his knowledge of everything.
Most of us know first the highways, and
THE CLASSICAL REVIEW. 61
then a few of the by-ways. Mr. Browning
used to know every inch of one highway
with all its associated by-ways, and never
set his foot on any other highway in the
same region. If he had beena zoologist, he
would have known all about lions and no-
thing about tigers. Of course, this is no dis-
paragement to his greatness. His true field
was not learning but life. Only, why could
he not have read some Plato? Our wistful
fancy cannot help framing some shadow of
the transformed Republic and interpreted
Phaedrus, for which we could have spared,
perhaps, the refutation of Bubb Doding-
ton and the divagations of the Famille
Miranda.
Besides interpreting Greek tragedy, he
translated it. The translations of the Alcestis
(in Balaustion) and the Hercules Furens (in
Aristophanes’s Apology) and the Agamemnon
are interesting, because they show us what
works he loved and what his theory of trans-
lation was. It is doubtful whether it
would have been a safe theory for any trans-
lator ; certainly it was not a safe theory for
Mr. Browning. ΤῸ translate word for word
and yet into beautiful verse, a translator
must have an ear filled and a mind pre-
occupied with minnte effects and delicate
experiments in language ; in short, he must
be something that Mr. Browning never was.
Moreover, he cared for the main matter too
much to satisfy this fastidious generation in
respect of ‘settling /oti’s business’ and
‘properly basing own.’ In short, these trans-
lations represent his play, not his work.
After the Agamenmon (1877) he did not
write on the Greek drama again. But the
impulse of his great Greek period remained.
His later volumes are full of both Greek
subjects and Greek references. One thing
that he has learned, from his own interpret-
ing of Euripides, is a new interest in myth-
ology. (‘A myth may teach. Only, who
better would expound it thus Must be Euri-
pides, not Aeschylus.’ Parleyings, Mandeville.)
Besides smaller references, there are the pro-
logue to the Parleyings, with its rehand-
ling of Admetus’s story, and Jaion (Jocoseria),
with its conversion of the transgressor
into a newer and more human Prometheus.
But also there isa word in Gerard de Lotresse
(Parleyings) for those who might think that
we must go back to myths for all our poetry.
Outlying stories that he found in his great
researches are worked up in Lchetlos and
Pheidippides. He is still constant to the
Aeschylus of his youth, and especially to
the Prometheus, and to these he adds Pindar
and Homer, quoting all three in Roman
letters in the midst of his verse. From
Homer he is led to consider the ‘ Homeric
question,’ and uses it characteristically to
show forth in allegory the religious edu-
cation of mankind (Asolando, Developments).
In this period, for the first time in his life,
he begins to add Latin to Greek. He ex-
pands three playful lines of Virgil into the
weird and pathetic mystery of Pan and Luna
(Dramatic Idylls, 11.). He refers three times
to Juvenal (once in Pacchiorotto, twice in
Pietro of Abano), (if I am right in so inter-
preting ‘Sylla cuts a figure, leaving olf
dictating,’ and ‘ while the half-mooned boot
we boast’). But his pre-eminent Latin poet
is Horace. Time after time he quotes him,
and incorporates his Latin into the verse
(‘ You’ve wine, manhood’s master! Well,
‘rectius si quid Novistis impertite!’ Wait
the event,’ where I cannot free my mind
from a gruesome suspicion of dmpertite).
The final fruit of his Latin reading 15
Imperante Augusto (Asolando), of which
nothing less can be said than that it might
stand in Men and Women. Whether the
subject-citizens of Augustus spoke exactly
like this is as indifferent as the question
whether Cardinal Wiseman defended himself
like Bishop Blougram, or Andrea del Sarto
felt himself dragged back from immortality
by his wife. We may not have the situa-
tion as the Romans conceived it, but we
have it as they would have conceived it, if
they had had a poet great enough to show
them how.
T. C. Snow.
To the Evrror of the CLASSICAL REVIEW.
DEAR SIR,
The enclosed lines, addressed to Mr. Robert Browning, and accompanying a little volume of Greek
verse, were very graciously received by him last summer, and may on that account have some interest for
your readers at the present time.
Ὦ paxape, ὅστις παντὸς ἀνθρώπου μαθὼν
ψυχήν τε καὶ φρόνημα καὶ γνώμην, λόγοις
ἐν ποικίλοις τε καὶ καλοῖς ἐφημίσω,
ὅσαις θ᾽ ὃ χρηστὸς φροντίσιν μάτην πονεῖ,
ὅ τ᾽ αὖ πανοῦργος ἀθλίως θυμοφθορεῖ,
LEWIs CAMPBELL.
χὠ μικτὸς ἀμφοῖν οἷον ἐξαντλεῖ βίον
γραφαῖς ἀριστ᾽ ἔδωκας ἐξῃκασμένον,
- ἄρ᾽ ἂν δέχοιο καινὰ δὴ μιμήματα
ἔργων παλαιῶν γνωρίμων τέ σοι τάδε,
μνημεῖ᾽ ἔμοιγε φιλτάτης ὁμιλίας ;
62 THE CLASSICAL REVIEW.
BISHOP LIGHTFOOT’S LITERARY WORK AT DURHAM.
Not many generations have passed since a
Bishopric was regarded as the natural goal
of a scholar’s life, in which, with much
leisure and a few occasional routine duties,
he might devote himself almost exclusively
to his studies. But those days have happily
passed away, and the present danger is rather
that in the demand for practical men of
business capacity the claims of scholarship
to be duly represented in the highest posts
of the Church should be passed over, The
calls upon a Bishop’s time and energies are
simply endless: the enormous growth of
population during the past twenty or thirty
years in most of the English dioceses has by
no means been met by a few tardy subdi-
visions; and the administrator of a large
diocese has his hands so full, that it seems
an impossibility for him to save any time for
literary work. Every parish demands indi-
vidual attention, and looks for at least
occasional personal visits; Confirmations
must be multiplied until they are held
annually within reach of every parish ; there
is not a Church Society but looks to the
Bishop to champion its cause in various
centres ; Conferences, Committees, Organi-
zations, Gatherings of special bodies, etc.
tend to increase on every side; while the
daily post alone brings in enough work to
employ the time of one man. In such a life
as this, how is it possible for a Bishop, how-
ever gifted, to secure any leisure for literary
work ?
It was such considerations as these that
aroused a serious anxiety in every quarter,
when the announcement appeared in the
newspapers on January 28, 1879 that
Professor Lightfoot had accepted the see of
Durham in succession to Dr. Baring. With
regard to the episcopal appointment, as such,
the news was greeted with universal sitis-
faction; but it was felt that the price
would be altogether too great to pay fer a
powerfui administrator in the Church if Dr.
Lightfoot’s new responsibilities should pre-
vent him from giving to the world any more
of the eagerly expected results of his life-
work on the writers of the New Testament
and the Apostolic Fathers. A scholar of
less note, it was urged, might well be found
to organize and guide even such a great and
difficult diocese as that of Durham; but no
one could fill Dr. Lightfoot’s place as a
teacher and an expositor.
In the three months which intervened be-
tween his appointment and his consecration,
while he was still at Cambridge, letters
kept continually pouring in upon the
Bishop-elect from all manner of correspond-
ents, imploring him to find time in some
way or another to continue his literary
labours ; and Dr. Westcott’s sermon at the
consecration of his friend in Westminster
Abbey, in which he sketched the ideal of a
Bishop’s work, contained an earnest plea for
patient thought and study and wise counsel
on deeper subjects than mere diocesan
detail or development. To one and all of
these appeals the Bishop himself returned
one steadfast answer ; ‘he had not accepted
the oversight of the diocese to neglect its
duties. Experience would show, but he
would not venture to predict, whether any
time would be left him to continue his
literary work.’
Accordingly from his first entering on his
new sphere he devoted himself unflinchingly
to the administration of his diocese; and
frequently for weeks or even months at a
time he found it impossible in the pressure
of other work to secure any leisure for
literary production. While however
throughout his episcopate his diocese held
the first and paramount position in Bishop
Lightfoot’s thought and energy, he consist-
ently kept before him as only a secondary
responsibility the urgent claim which rested
upon him as a scholar and a theologian to
strive earnestly to finish the work which he
had undertaken before he became a Bishop.
It was this constant sense of a great duty
incumbent upon him that led him to devote
every leisure hour that could be spared
from diocesan work to the prosecution of his
literary labours.
It is not an easy matter to point to any
definite time or occasion which the Bishop
was able regularly to secure for his books in
the midst of his busy life at Auckland
Castle. In the earlier years of his life there
his habit was to rise very early in the
morning, and lighting his own fire (which
had been laid ready for him over-night) to
make sure of two or three hours’ quiet work
in his bedroom before breakfast. But after
a few years when the terrible strain that
pressed upon him began to tell upon his
health, he reluctantly abandoned this plan
as anything like a general rule.
When his constant engagements took him
from home, he would sacrifice any personal
convenience to return before night, or at
least very early the following morning, in
THE CLASSICAL REVIEW. 63
order to save as much time as possible. But
even so the days at Auckland were seriously
brokeninto. After breakfast he went through
his letters with his chaplains, reserving a
certain number to answer with his own hand,
amongst which were the numerous com-
munications he constantly received from
scholars in all parts of Europe. The pre-
paration of sermons, speeches, charges, etc.,
necessarily occupied a great deal of attention.
And though the position of Bishop Auckland
saved him from a large number of the
inconsiderate callers who, had he lived at
Durham, would have occupied his time
about matters that could have been dealt
with as well by post, still there were not a
few to whom an interview was really im-
portant, and who accordingly found their
way to Auckland Castle. But with all these
interruptions the last hour or two of the
morning not unfrequently found the Bishop
engaged with his literary work, and he was
often able to keep the greater part of the
evening for it. Unfortunately too he would
day after day restrict his exercise to a short
stroll in his park, and then return to his
work for the rest of the afternoon. For a
man who had been used to a considerable
amount of walking, this loss of fresh air and
exercise was a serious strain upon his
health.
The habit which the Bishop had formed of
turning to his books at every available
opportunity, however short, was exemplified
even in the smallest details. Thus on his
constant railway journeys, or in his long
drives to the outlying villages of his diocese,
he always had with him as his constant
companion ‘a bag (familiarly known as ‘the
Pandect’), in which were ready to his hand
books, literary periodicals, proof-sheets, etc.,
for reference at any spare moment.
There was however one great opportunity
for uninterrupted work open to the Bishop,
which he was not slow to seize. When
August came round, and he was able to get
away for a summer holiday, he would carry
off his books to some retired spot—gener-
ally in Scotland, and by preference to
Braemar, where the bracing air and the quiet
enabled him to work freely—and there he
would abandon himself once more to a
student’s life. His diocesan correspondence
followed him even there, but it did not reach
him until mid-day, and the mornings and
most of the evenings were kept sacred for
literary work. It was during these holidays
that a great part of his introduction to the
Ignatian Epistles was written.
In the great bulk of his literary work
Bishop Lightfoot depended entirely on his
own labours. He never employed an aman-
uensis ; he rarely allowed any one else even
to verify his references. The only relief
which he would accept was the almost
mechanical correction of the proof-sheets of
the new editions, as they were called for,
of his ‘ Epistles of 5. Paul.’ But latterly
he entrusted more and more of his editing
work to his chaplain, the Rev. J. R. Harmer,
who had prepared the indices for the edition
of 8. Ignatius. In passing the sheets of
his books through the press the Bishop
spared no pains to ensure completeness in
every detail ; thus, for instance, one sheet of
Ignatius was kept back for months to enable
him to add if possible an English rendering
which would preserve the play upon words
in κακοδαίμων in the Antiochene Acts of
Martyrdom of 8. Ignatius (δ 11).
One great secret of the Bishop’s being able
to produce such a monument of learning and
research as his Ignatius in the midst of an
exceptionally active episcopate was the
unique store of knowledge which he
brought with him from Cambridge, and the
remarkable accuracy of memory which en-
abled him to apply it readily. Page after
page was written currente calamo with
few or no books of reference at hand, and
with only a ‘ver.’ here and there in the
margin where future verification was re-
quired. He also had ina marked degree the
power of again taking up the thread of his
work after an interruption without a mo-
ment’s hesitation. The thought of his
complete and minute command of the whole
range of the first three centuries excites a
keen regret that the pressure of other busi-
ness in the first instance, and afterwards the
state of his health, should have prevented
him from carrying out his original project,
of writing a full historical introduction to
his articles on ‘Supernatural Religion’ before
re-issuing them in book form.
It would however be an inaccuracy to
imply that all the Bishop’s interest and re-
searches were confined solely to the period
of the Karly Fathers. Apart from the
various topics of general and current interest
which engaged his attention, he was a tho-
rough enthusiast and expert on the subject of
English Church history and antiquities,
especially with regard to the unique heritage
of his own diocese. He was among the first
to claim for the Northumbrian mission of
the seventh century its true position in the
evangelization of England; and he was
θ4 : THE CLASSICAL REVIEW.
familiar with every detail of the ecclesiastical
antiquities of Durham and Northumberland.
Auckland Castle came into his hands with
few or no relics of the see: he left it a
monument of the history of all his great
predecessors from the days of Aidan him-
self. Stained glass windows, shields, epis-
copal seals, portraits, books, personal relics—
such as the one faulty inscription of Butler,
or the desk of Cosin—all tell the story of
the past. In the summer of 1886 he began
to prepare a monograph on the history of
Auckland Castle, at which he continued to
work to the end as occasion offered. His
sermons on the north country saints,
preached at various churches dedicated in
their names, will form a series which in
point of Early English Church history will
carry far more than a merely temporary or
local interest. The Bishop himself intended
to publish these in a collected form when he
had completed the whole cycle, according to
the plan he had laid down for himself.
When Dr. Lightfoot was appointed to
Durham in 1879 there was some hope that
he might be able to continue at intervals to
give at the northern University some of the
lectures on the Greek Testament which had
made him so famous as a teacher at Cam-
bridge. His official position as Visitor of
the University and Patron of the Canonries,
with which two of its Professorships are
endowed, seemed to give a certain ground
for asking this of him. On more than one
occasion during the first two years of his
residence in the north he was urged to
undertake such a course of lectures. But
nothing would induce him to accede to this
request. He felt that his hands were more
than full of work in other directions, while
the teaching staff of the University was
amply sufficient for its needs. His interest
however in the University never flagged,
and it found a practical expression in the
foundation by him in 1882 of the De Bury
Scholarship for students who intend to take
Holy Orders in the diocese of Durham.
In his own home at Auckland Castle he
gathered round him a band of graduates of
the older Universities, who were reading
with a view to taking Holy Orders in his
diocese. The teaching of these students was
entrusted chiefly to the resident chaplains, of
whom there were always two on account of
this special work. The Bishop himself occa-
sionally gave them a course of Greek Testa-
ment lectures, and the general direction of
their studies rested with him: but more
than this he was unable to do. Altogether,
in the ten years eighty of these students
have been trained at Auckland.
The one impression left upon the minds of
all who knew Bishop Lightfoot, on a review
of his ten years’ episcopate, must always be
that of a Father of the Church, who set him-
self to rule over his diocese with conspicuous
devotion, judgment, and ability; whose
power of work seemed to be without limit,
whose liberality was without stint; the
motto of whose life was to spend and be
spent for those to whom, as he himself ex-
pressed it on the day of his enthronement at
Durham, he had given himself wholly for
better or worse. And when to all his other
labours was added the strain of the Lambeth
Conference of 1888, in which he bore no
small part, it was the last burden which
hopelessly broke down his already overtaxed
strength. In the midst of a life of such
ceaseless and varied activity, it was only by
the stern exercise of his inflexible will, and
a steadfast and self-denying earnestness of
purpose, that he was able in any degree to
continue his literary labours.
H. E. SAvaGeE.
‘He was strong by that sobriety of judg-
ment which commands the old, and that fire
of enthusiasm which wins the young. His
interest centred in the fulness of human
life. Speculation had comparatively little
attraction for him. Even this limitation of
his intellectual character increased his influ-
ence and effectiveness in dealing with con-
crete facts, He shrank from indistinctness
and indecision. Nothing visionary, nothing
that men call mystical, marred the effect of
his masculine reasoning ... In argument
and in exposition he preserved a true sense
of proportion. His learning was always an
instrument and not an end. No investigation
of detail ever diverted his attention from
the main issue. He mastered two outlying
languages, Armenian and Coptic, in order to
deal more surely with the secondary mate-
rials of the Ignatian controversy, but no
ordinary reader would know the fact. For
him the interpretation of ancient texts was
a study in life. He held books to be a witness
of something far greater, through which
alone they could be understood. A Greek
play or a fragmentary inscription, or a letter
of Basil or a homily of Chrysostom, was to
him a revelation of men stirred by like
passions with ourselves, intelligible only
THE CLASSICAL REVIEW. 65
through a vital apprehension of the circum-
stances under which they were written. He
was a born historian. ‘“ How I long,” he
said to me more than once, ‘‘ to write a his-
tory of the fourth century.” If he has not
written it, he has shown how it must be
written.’
[Sermon preached at Westminster Abbey by
Canon Westcott on the Sunday following the
Bishop's death.
The interesting words just quoted will
only deepen the feeling of regret with which,
if we are not mistaken, scholars in general
regarded the self-sacrificing obedience of
Prof. Lightfoot to the call which removed
him from a centre of thought and learning to
the distractions of a busy and populous dio-
cese. Granted that his influence as Bishop
was of the highest value to the Church
at large and to his own diocese in particu-
lar, we have to set against this the heavy
loss to Cambridge, which under the fostering
care of himself and his brother professors
had become perhaps the most important
seed-plot of the Church in England ; and,
besides this, there is the thought of the
added years of fruitful literary work which
might have been hoped for if he had con-
tinued to reside in his University. Was it
really necessary that our Beda should be
forced to take up the work of a Ceadda as
well? However the past cannot be undone.
It is some consolation that, as we are told,
there is a prospect of a rich aftermath, in-
cluding a much-enlarged edition of Clement,
several volumes of sermons and addresses, an
unfinished volume on the Northumbrian
Saints, notes on the remaining Epistles of
St. Paul, and possibly some notes on
Aeschylus.—Ep.
ARCHAEOLOGY.
ROMAN INSCRIPTIONS FROM
SARDINIA.—II.
Signor Tamponi, the energetic Sardinian
archaeologist who recently communicated to
the Classical Review (111. 228-233) an im-
portant find of milestones, has conferred a
fresh favour by sending the inscriptions
printed below, all of which, as I understand,
are unpublished. At the editor’s request I
have added, as before, a few notes.
BK. HAvERFIELD.
ITI.—MILEsTONES.
The following belong to the same series as
those published in the Wotizie (1888 p. 535)
and in this Review (p. 228), being found near
Olbia (Terranova), nos. 1-4 at Sbrangatu, no.
5 at Telti, on the Roman road to Carales
(Cagliart). All are fragmentary, and are
interesting mainly as being unpublished. The
supplements in italics are inserted (conjec-
turally) for the reader’s convenience only.
The details given in italics are quoted from
Sig. Tamponi.
1. Height 32 in., circumference 42 in.
MP) CLX., {II
MIVS
PIVS * FELIX Aug
PONT * MAXIMYS
NO -pexexex, VOL. IVs
5 TRIB * POTESTATE
viam quae DVCET A KARAL * OLBIE
VETVSTATE CORRVPTA72
restituit CVRANTE Μ᾿ PI
vs
10 Wee
The emperor’s name, if the text be right,
must be Septimius (Severus or Geta). The
procurator’s name cannot be restored: no
known governor’s name begins 77. Pz. . .,
nor, indeed, are nomina beginning in δὲ at
all common.
2. Height 37 in., width 15 in., thickness
8 in.
M P CLXVIIII
DN IMP CAES
FLABIO
CAES
FEL * INVIcto
ΓΙ}
1... iid i KARALIBVS
OLBIAE * BETVState corruptam
RESTITV2t
Obviously one of the Constantines.
3. Height 52 in., width 16 in., thickness
12 an.
F
66 THE CLASSICAL REVIEW.
| Pel Siig, i
QVAE DVCIT
KARALIBVS * OLB
BETVSTATE * COR
REST
ΡΣ fefoky
4, Height 44 in., circumference 16 in.
5. Height 78 in., circumference 56 in.
M P CLXV
DN IMP CAES DOM
ANVS
E EPIRA ΜΝ
5 KARALIB * OLBI
CORRVPTA Restitutt
PRNETIVS VP
PRESES Provinciae
Sardinte
7. Vir) P(erfectissimus)
In line 2 the supplement Dom[iti]anus is
impossible, because (1) the title d(ominus)
n(oster) appears first in 194 a.p. (2) Lmp.
Caesar is unusual till 100 a.p. (3) vir per-
Jectissimus is unknown till 200 a.p. (Hirsch-
feld, Verwaltungsgesch. 274 n. 1). (4) Under
Domitian Sardinia was governed by a sena-
torial magistrate. No doubt Z]|Domlitius
Aurelijanus is meant. The procurator of
Sardinia was sometimes called praeses even
before Diocletian.
TV.—InstRvYMENTVM DoMESTICVM.
Since the editors of the Corpus led the way,
archaeologists have begun to pay great atten-
tion to the smaller inscribed objects, such as
bricks, lamps, pottery, which earlier scholars
except perhaps in England had grievously
neglected. There is, indeed, a good deal to
be learnt from these minutiae. We do not
usually think of the emperors as wholesale
manutacturers. But the Julio-Claudian
family owned in the officina Pansiana a brick
and earthenware factory which’ supplied
all Italy and even Dalmatia, while Marcus
Aurelius and his mother Lucilla before him
derived a princely revenue from their brick-
yards. We do not, again, usually imagine
that means of transport were sufficiently
advanced in the first two centuries A.D.
to enable capitalists to supply distant regions.
But the lamps of almost the whole empire
were probably made in Italy, while the red
‘Samian’ ware which was universally used,
e.g. in Roman Britain, was all but entirely
imported from certain potteries in France.
A Flat tiles.
1. Circular stamp 3 in. in diameter, with
a figure of Victory tol. Beneath in a
semicircle. STATI . Μ΄. LYCIFERI
Compare in, C.J.L. vol. 5,
8048 388. Statius Marcius Lucifer fec(it)
89. MM. Stati Luciferi
On mortaria (pe/ves) from Pompeii and
Herculaneum. The same factory no doubt
made all the rougher kinds of ‘ baked earth’
ware (opus doliare).
Compare also, in C.L.L. 5,
8048 34 35 St. 17. Celer (mortaria, Pom-
peli, etc.)
8048 Τ᾽ = Stat. Marcius, Demetrius fee.
(tiles, Antium, Velitrae)
8056 294 Stat. Ma. ..., Demetr. .
(pottery, Puteoli)
8048 4° Statius Marcius, Primigen|i|us
fec. (mortarium, Stabiae)
ibn (so mi, =)
Sec. in figlin|is|
8042 (so 8043 65)
(tiles, Antium &c.)
8048 45. St. Marcius, Tognaeu|s| f (mort-
aria, Pompeii)
More might be quoted, but these
are enough to show that we have a firm
using two nomina, Statius and Marcius, and
a varying cognomen. Descemet (Jnscriptions
Doliares un. 223-5) includes the work of
some Statii Marcii among the products of
the ‘ Domitia Lucilla’ works (circ. 123 a.p.),
but only because the formula Valeat qui fecit
appears on both sets of stamps.
2. M * LOLLI
TIRACAES
This seems to complete C./.Z. 5, 8046 29
and Ὁ,
Stat. Marcius Secun|dus |
(mortaria, Pompeii)
St. Mare, Stator f
3. ZIMPONI
Probably imperfect.
4, HEREN
Tiles with this name are not uncommon.
ὃ: EKO) ἨΔ agit
6. anst C.I.L. 5, 8042 19,
7. SILV ° CIS
8.
Round a semicircle.
© ‘ LICINI DONA. ...
THE CLASSICAL REVIEW. 67
B. Lamps, with reliefs representing flowers.
9. BIC AGAT
6.1.1. 8053 45. (Cagliari).
10. L * CAESSAE
Read caz[c].
11. AVFIFRON
Both these are common lamps (Wilmanns
2832), and have been found even in Britain
where inscribed lamps are relatively uncom-
mon; the Guildhall museum contains an
unpublished ex. of each (of the latter the
variety AVFFRON), and Mr. Clayton has one
of the latter at Chesters.
C. Glass.
12. On the bottom of a glass vase, with a
female figure.
RIMON
Probably incomplete. Cp. 6.1.1. 12 5696,!9
Patrimonium and Patrimgni (sic), and a
fragment found and preserved at York
(museum catal. p. 90) PpaTRIM.... (wrongly
read and published by Mr. Th. Watkin.)
D. 13 A gold ring with a precious stone, show-
ing the seal :—
PV
VI‘ FE
pv are the initials of donor or recipient.
VT. FE=utere felix, as common a formula as
the Christian vivas in deo.
E. Pottery (smaller) :—
14, RASINISPIS
Ο.1.1. 5. 8056 300 % 203, Pasint Pisani
(potter’s name).
15. ... Lis 16 .. ANVS
17 Among the ornament apparently out-
side
4Va (LERIO)
17—22 cylindrical vases
Jlowers : on the foot ;—
decorated with
ΤΠ OE Sg. Ex ΜΡ
Sy Ne One 205° ‘sic
21. MM 5. 22. ACVIVS
24, On the handle of a vase :—
OPID
Sig. Tamponi sent drawings of Nos. 1, 8,
17, which I have tried to represent by
description.
Ἐς HAVERFIELD.
Tier- und Phanzenbilder auf Miinzen und Gemmen
des klassischen Alterthums. By F. ImuHoor-
BiuMER and Orro KELLER. Leipzig, 1889. 4to.
24 Mk.
THIs most useful and long-needed work consists of
twenty-six large photographie plates of coins and
gems representing animals and plants, and of 161
pages of descriptive text, with full indexes,
The value of numismatic and glyptic illustrations
of the natural history of the ancients is well-known,
but hitherto the inquirer has had to select his coins
and gems from miscellaneous works and periodicals.
The numismatic portion of the present book has been
prepared by Dr. Imhoof-Blumer with his usual sound
judgment and wide knowledge. The coins selected
are principally Greek of the autonomous period. The
text gives a full description of every coin illustrated,
with a mention of the cabinet where the specimen
exists. It is perhaps a little to be regretted that the
approximate date of the coins is not stated, especially
for the benefit of those who are not numismatists. A
reference to Mr. Head’s Historia Nwmorum will,
however, in most cases enable the student to gather
this information for himself. Dr. Otto Keller (the
author of the Tiere des Klassischen Alterthwms) has
most carefully brought together a very large selection
of ancient gems engraved with subjects from natural
history, and has written the accompanying text.
With regard to the general arrangement, the coins
and gems form separate sections, and in each an
alphabetical order of subjects is adopted. The
mammalia stand first, and then follow the birds,
reptiles, fish, insects, plants, &c., nor are even the
‘ Fabelhafte Tiere’ and ‘ Mischwesen’ forgotten.
WARWICK WROTH.
Manual of Ancient Sculpture. By PIERRE PARIS.
Edited and augmented by JANE E. HARRISON.
London, H. Grevel and Co., 1890. 8vo. pp. xvi.,
369 ; 187 illustrations. 10s. 6d.
This is a volume of the ‘Bibliotheque de l’En-
seignement des Beaux-arts,’ and in the English
version it belongs to the ‘Student’s Fine Art
Library.’ The aim of the book is to supply not a
history of sculpture, but a running commentary on
the artistic and aesthetic qualities of selected
specimens. Exception may well be taken to the
whole scheme of the book. Facts and dates
are for the most part shadowy, but the student
is given an elaborate series of aesthetic principles.
It seems a more profitable plan, to supply the
framework of fact, and to suggest the principles
involved in the tersest language possible, leaving
the reader to develop the details, as far as his
powers will allow.
If however the author’s point of view is accepted,
the work is fairly well performed. The commentary
is as readable as can be expected; it adopts, as
a rule, the accepted, not to say the conventional
views ; and there is an abundance of illustrations.
The hand of the English editor is visible through-
out the Greek and Roman sections. Short. biblio-
graphies (which would be better, if they were
more accurately printed) are prefixed to each
section ; distinct errors have in several places been
corrected ; more specimens have been quoted.
Allusions have been introduced to recently pro-
pounded theories, about which M. Paris is silent,
either by design, or because he has not followed
the latest archaeological literature. More than
once, also, there are hints that the editor has a
more personal knowledge of Greece than the author.
It should be added that in some instances the
artistic judgments have been almost re-written,
although the editor’s preface suggests a contrary
impression. Compare the criticisms of the Caryatids
of the Erechtheion (p. 235) and of the sculptured
drum from Ephesus (p. 267). The illustrations
have been increased in number, and occasionally
improved in quality.
10 is to be regretted that the translation, which
is in many ways an improvement on the original,
F 2
68 THE CLASSICAL REVIEW.
is disfigured with a great number of misprints.
Some are obvious, and others are more obscure as
Caovoclias for Cavvadias on p. xvi. I conjecture
that ‘chit’ on p. 159 should be ‘club’; that on
p. 184 ‘Berlin Museum’ should be ‘ British
Museum’; that on p. 261 the ‘Doryphoros in
the British Museum’ should be the ‘ Diadumenos
of Mr. Blacker.’ On p. 858 Kanachos should be
read for Kalamis. Pasiteles appears on p. 184 as
Praxiteles, and on page 366 as Pastiles. On p. 317
the English version assigns to Lessing a theory
that Titus was the sculptor of the Laocoon, which,
as is justly observed, ‘is of course impossible.’
A. Η. SMITH.
The Horsemen of Tarentum. By Arruur J.
Evans, M.A., F.S.A. London, (Quaritch), 1889,
8vo. 12s. 6d.
This most able and thorough monograph on the
coins of Tarentum with the horsemen types was
originally printed by Mr. Evans in the Numismatic
Chronicle, and is now, with the addition of indices
and a table of contents, very welcome in the more
accessible form of a separate publication. It is very
fully illustrated by eleven autotype plates and by
some woodeuts. The coins are distributed into ten
periods (B.c. 450-209), the description of the money
of each period being preceded by an introduction.
A general introduction occupies pages 1-29. The
chronology of the different issues is worked out with
minute care, and the arrangement proposed certainly
seems convincing. Some important evidence is
derived from various hoards of Tarentine coins
examined by Mr. Evans himself. In several of the
varieties of the horsemen and Taras types, Mr.
Evans ingeniously finds allusions to the actual
history of Tarentum, and it can hardly be doubted
that such allusions exist. There is an excellent
section on engravers’ names. Mr. Evans points
out that the Tarentine engravers, some of whom
certainly worked for other cities of Magna Graecia,
signed the coins not only in their artistic capacity
but as officials responsible for the weight and metals.
The monograph is one which all numismatists will
study with the greatest interest, and it will well
repay the attention of classical archaeologists, in
spite of the minute treatment necessarily entailed
by so complex a series of coins. Mr. Evans should
not fail to give us similar essays on the coinages of
other cities of Magna Graecia,
Warwick Wroru.
A Guide to the Principal Gold and Silver Coins of
the Ancients (B.c. 700-a.D. 1). By Barcuay V.
HEAD, D.C.L., Ph.D. London, 1889, 8vo., with
7 plates. Third edition. 2s. 6d.
This is a new edition of the admirable guide
written by Mr. Head to accompany the large series
of electrotypes of Greek and Roman coins selected
by him for exhibition in the British Museum and
other public institutions. The work is now so well
known that it will be enough to call attention to
the fact that the author has in the present edition
(the third) made many material improvements in
accordance with his Historia Numorwm and other
numismatic publications that have appeared since
1881, the date of the second edition of the Guide,
The grouping of the coins in periods remains,
however, almost unaltered and the arrangement of
the plates has, very wisely, not been disturbed. The
half-crown edition now before us is accompanied by
Seven autotype plates giving a selection of some
of the most interesting coins of each period. The
edition of the Guide costing twenty-five shillings
and illustrated by seventy plates is not yet out of
print, but will, when re-issued, be accompanied by
the same revised text as the new half-crown edition.
δ ἘΝ:
ATHENAIOS, xi. 49.—In describing the type of
drinking-cup termed καρχήσιον, Athenaios quotes the
following description of it from Callixenos of Rhodes :
ποτήριόν ἐστιν ἐπίμηκες, συνηγμένον εἰς μέσον ἐπιεικῶς,
ὦτα ἔχον μέχρι τοῦ πυθμένος καθήκοντας. He then
observes that the mast-head of a ship was also termed
καρχήσιον, and that Asclepiades of Myrleia said that
the drinking-cup was copied from the mast-head. <Ac-
cordingly he quotes from Asclepiades a description of
a mast containing the following description of the
mast-head : ἔχει δὲ τοῦτο κεραίας ἄνω συννευούσας ἐφ᾽
ἑκάτερα τὰ μέρη, καὶ ἐπίκειται τὸ λεγόμενον αὐτῷ θωρά-
κιον, τετράγωνον πάντῃ πλὴν τῆς βάσεως καὶ τῆς
κορυφῆς" αὗται δὲ προὔχουσι μικρὸν ἐπ᾽ εὐθείας ἐξωτέρω.
This is repeated by Eustathios, p. 1423, but more
briefly : καρχήσιον, ᾧ ἐπίκειται τὸ λεγόμενον θωράκιον.
The whole passage is translated by Macrobius, Satwr-
nalia, v. 21,5: but his translation is useless; for
he has mistaken ἱστοῦ for ἱστίου, and thus gone
hopelessly astray.
The top and bottom of the θωράκιον, which were
not τετράγωνος, bulged slightly outward : therefore
the intervening portion, which was τετράγωνος, did
not bulge; in other words, it had straight sides.
This seems to be the whole meaning of τετράγωνος in
the curious phrase κύλινδρος τετράγωνος πανταχεῖ,
in Corp. Inscr. Attic. vol. ii. no. 835, 1. 70: for the
object would hardly be termed κύλινδρος if its trans-
verse section were square. As the θωράκιον was the
crow’s-nest, the transverse section would more pro-
bably be round than square. Now the καρχήσιον
(the drinking-cup) is described as elongated and con-
tracting gently in the middle: and the transverse
section of every drinking-cup was necessarily round.
Thus the body of the καρχήσιον (the drinking-cup) so
far corresponds in shape to the θωράκιον, that I
would venture to read ἔγκειται for ἐπίκειται, so as to
make the θωράκιον the central portion of the καρχή-
σιον (the mast-head) : the horns curving up on each
side of this (κεραίας ἄνω συννευούσας ἐφ᾽ ἑκάτερα τὰ
μέρη), consequently corresponding to the handles
reaching down to the bottom (ὦτα μέχρι τοῦ πυθμένος
καθήκοντα) of the drinking-cup. As these κεραῖαι are
described as συννεύουσαι, they clearly cannot be
κεραῖαι in the usual nautical sense; for then they
would be straight. A pair of rings for the halliards
is sometimes plainly represented at the mast-head, as
in the Corinthian painting in Antike Denkmdler, vol.
i. pl. 8, fg. 3; and these resemble the handles of a
drinking-cup. But such rings would not necessarily
be perfect ; for the lower part would alone suffice to
hold the halliards, so that the upper part might
remain open: and then they would resemble a pair
of horns. The corruption of ἔγκειται into ἐπίκειται 15
highly probable: for Asclepiades describes the mast
from the bottom upwards to the top ; so that a copy-
ist might easily conclude that, as the θωράκιον is
mentioned after the καρχήσιον, the one was really
above the other. Ceci ToRR.
PAUSANIAS v. 11, 1.—Pausanias in his descrip-
tion of the Zeus Olympios of Pheidias (v. 11, 1)
says that his himation was adorned with various
animal forms and lilies: and in v. 22 he describes a
statue of Zeus, probably of the pre-Pheidian period, by
the Aeginetan Aristonoos, on the head of which was
a crown of lilies. Whether lilies appear elsewhere as
an attribute or symbol in the representations of Zeus
I do not know. It is usual to say that Pheidias only
THE CLASSICAL REVIEW. 69
used them as symbols of beauty in general. But a
passage in Athenaeus (15, p. 684 D) suggests a
more specially appropriate interpretation : ‘Nixav5pds
φησιν ἐξ ἀνδριάντος τῆς κεφαλῆς ᾿Αλεξάνδρου τὴν
καλουμένην ἀμβροσίαν φύεσθαι ἐν Κῷ."...τὸ κρίνον οὕτω
λέγουσι. It may be that the legend arose from the
record of a crown of lilies carved on the head of Alex-
ander’s statue : in any case it is clear that the mean-
ing of the passage turns on the word ἀμβροσία, and
the idea is that the lily expressed the immortality or
apotheosis of Alexander. It may well then have the
same meaning on the himation of the Olympian Zeus,
and on the head of the Aeginetan work they may be
an allusion to the ἀμβρόσιαι χαῖται.
L. R. FARNELL.
PINDAR, PyTH. xii. 16, εὐπαράου κρᾶτα συλάσαις
Medoloas. εὐπάραος has been generally regarded as
a literary epithet, quite against the method of
Pindar. Is it not rather that the poet idealizes a
process which he saw going on around him in art ὃ
The type of Medusa on the coins of Coroneia was
gradually refined, and as early as 450 B.c. presents
a face which, but for the conventional lolling
tongue, has nothing of the goblin left [v. Brit. Mus.
Cat. Gr. C. Centr. Gr. pl. vii. 6 and note the
continuance of the refinement in pl. vii. 8, 385 B.c.
and vil. 9, 335 B.c.]. The twelfth Pythian was written
for Midas of Agrigentum, and in Sicily also a milder
type of Medusa’s ask was known [as at Kamarina,
Syracuse, Motye, Egesta: v. B. M, Cat. Sicily, pp.
39-40: 244-5: and Torremuzza Sic. Vet. Num. tab.
68, 19 and 20: though some of these coins are ivth
century]. On the other hand when Pindar, asin the
xth Pythian, writes fora Spartan audience, Medusa is
still encircled with a ‘horror of serpents.’ The epithet
εὐπάραος taken literally represents the one title to
beauty which the Medusa might be held to possess :
the full cheek, when once, as in the Coroneian coins,
the rictus of the monster disappears, is in truth
‘comely.’ Pindar, with the privilege of a Greek
poet, creates mythology by indicating the possibilities
inherent in the spirit of a contemporary art, which
again was indebted for its technical tradition to an
old-time half-identification of Medusa with Hathor.
Woop BEAMS IN STONE ARCHITECTURE. — In
the last number of the Philologisches Wochen-
blatt (2nd Nov.) Dorpfeld notes in reference to M.
Tsountas’ latest report on Mycenae that the walls
there mentioned are constructed with a balk of tim-
ber in between the courses of squared stone. Natu-
rally Dorpfeld recalls Solomon’s buildings [1 Kings
vii. 2]. To this reference however I would add
Pliny, V.H. xvi. 79, who states that the cedar-beams
in the walls of the Apollo temple at Utica had lasted
down to his own time. Similarly the Lycian archi-
tecture shows unmistakable traces of having formerly
employed beams of wood, where afterwards marble
was used: but in Lycia the case is somewhat different,
as building in stone seems to have been copied en bloc
from building in wood. In Africa and Palestine the
use of wood beams in alternation with stone is an in-
troduction of the Phoenicians : in Greece and Lycia
there may possibly be the influence of the same
master-builders. Phoenicia was a great building
nation, and its houses were famous [Jos. B.J. 1].
18-9: Ezek. xxvi. 12, where the construction with
stone and timber alternating seems to be directly
noticed : 1 K. vi. 36: Ez. xxvii. 3]. Certain pas-
sages in the Talmud also allude to a distinctive
Phoenician style of architecture, and there again the
element of difference may be the employment of
cedar-beams [Mishna Maaserath III. 5, an ‘ Atrium
Tyrium’: Baba Bathra III. 6, a ‘scala Tyria,’ and
‘fenestra Tyria’]. H. ARNOLD TubBs.
WAS THE PATERA USED IN SACRIFICES BY THE
Roman Rire ?—Baumeister in his Denkméler (fig.
1304) gives without demur—perhaps he is right—the
figure of a priest with veiled head (and therefore
sacrificing ritu Romano), who holds a patera in his
right hand ; and indeed familiar passages in Horace
and elsewhere naturally dispose us to associate the
patera with libations at Rome of any kind ; with this
too Marquardt seems to agree on page 654 of his
Privatleben, But for all that, questions suggest
themselves—what parts did the simpulum (or simpu-
vium) play, of which Festus says ‘quo vinum in
sacrificiis libatur’? And still more, how can we ac-
count for the fact that, whereas the patera was un-
questionably often made of silver, there is no mention
of it in Liv. xxvi. 36, along with the salinum and pa-
tella among the silver articles retained for religious ser-
vices in the year 210 B.c.? The easiest answer, if it
could be made good, would be that the patera, as a
Roman equivalent for φιάλη, came in later from the
Greek cities of Italy, and with the Graecus ritus.
It is true that the Etruscans had a vessel like a φιάλη
with a handle, which we usually call a patera, but
that differed from the ordinary patera (which was
precisely like the φιάλη), and we have no evidence of
its use at Rome. If we set down the patera as of
comparatively late introduction, perhaps not before
the latter part ot the 3rd century B.c., we may sup-
pose the old Roman vessels for libation to have been
the simpulum alone or the simpulum combined with
the capis; and these, with the conservatism of relig-
ion, would be retained whenever the ritus Romanus
was used.
We can thus understand why in the relief on the Ar-
cus Argentariorum, for instance, among the sacrificial
implements the simpulum and capis appear, but no
patera (if we suppose it to be a representation of
implements for the Roman rite): and why again the
simpulum is the badge of the old Roman order of
Augurs, while the patera marks the Epulones who
date only from B.c. 196. That the simpulum was
one of the primitive sacred vessels is clear from Varro’s
statement (Z.L. v. 124) that it had in old times been
used at the dinner table, but, being superseded there
by the Greek cyathus, was retained for sacrifices,
and from the passage in Juvenal (vi. 343), where it
is coupled with the ‘niger catinus’ ( ?=praefericu-
lum), which was probably for public sacrifices what
the smaller patella was for domestic. The term sim-
pulatrices applied to women would hence naturally
convey the imputation, if it is to be so regarded, of
old-fashioned piety. The omission of the simpulum
from the list of silver plate in Liv. xxvi. 36 is ex-
plained by Pliny’s remark (H.W. xxxv. § 158) that
the primitive earthenware for the simpulum was re-
tained, from a conservative sentiment, even in his
time. The words of Festus and Pliny seem (though
not beyond a doubt) to assert that the wine taken
from the larger vessel or crater was poured directly
from the simpulum, but in some cases the simpulum
must have been used to fill the capis, from which
the libation was then poured (cf. Liv. x. 7). The
view that the patera did not belong to the Roman rite
may be at any moment (and perhaps has been) over-
thrown by monumental evidence. The figure from
the Vatican, alluded to above, as shown in Bau-
meister’s Denkmdler, would overthrow it, if the patera
in the right hand of the veiled priest were authentic ;
but, as both hands are admitted to be restorations,
it is impossible in this instance to say what the sculptor
meant them to hold. G. E. MARINDIN.
70 THE CLASSICAL REVIEW.
MARBLE CALATHOS-STELE FROM Pout.—One inter-
esting product of this year’s excavations at Poli was
the worked stone found in grave K 29. There can
be no doubt that it is a calathos, and, so faras I
know, such a sepulchral monument is unique. The
stone was apparently set up on a stele of some
sort, for on its under side is a fusing-hole :—from
tomb K 2 came the counterpart, an octagonal stele
with a similar hole above. A reproduction of such a
tombstone may be seen for example on a Campanian
yase in the Berlin collection [No. 3317]!; and the
practice of making a calathos the actual grave-monu-
ment is the natural outeome of the custom of which
Vitruvius speaks in the well-known story about the
origin of the Corinthian—or calathos—capital in
Greek architecture.* In itself the calathos as a
basket is simply a useful piece of household furniture,
but its connection with funeral rites lends it a certain
specific character as the appropriate vehicle of offerings
to the dead. In this sense it is frequently placed, by
S. Italian vase-painters, inside the aedicula itself.
The border-line between this use of the calathos and
that in which it is merely an article of furniture is
not very clearly marked, so that one hesitates to
assign a symbolic meaning to its presence on 6.4. the
Albani relief, or ar. f. hydria in Berlin, where the
sevne is sepulchral.4 It is just possible that in the
light of the stone found at Poli, and of vases like
those in Tischbein (iv. 47 and v. 7), we should
recognize a ‘sacrifice at the grave’ in a series of
scenes where a calathos occupies the centre, as though
it had been an altar, to which libations and other
offerings seem to be made: this series is mainly con-
fined to the small Athenian τ. f. lekythoi of humble
workmanship which date fron the fourth century. It
seems at least probable that the scenes on these vases
were intended to have a funeral significance ; and
where they are taken by themselves little more than
genre, there is still a suggested connection between
the life here and that beyond the grave.
The contents of tomb K 29 at Poli cannot be dated
with entire certainty, For the most part they belong
to the middle of the fourth century ; but it is not
precisely determinable whether the tomb has not been
used twice, at which indeed the find-spot of the
calathos rather hints. The stone was lying in the
1 Cf. No. 3024 zbid.
2 Vitr. De Arch. IV. i. 9. sqq.
3 ¢.g. Berlin No. 3171, or Millin JI. 29.
4 Antigq. No. 2379. Furtwingler gives a strange
explanation as though the scene had some connection
with the Paris story. The κάλαθος here indicates the
γυναικωνῖτις, Whither Hermes has come as ψυχοπόμπος
to carry off the wife and mother to the shades. That
this is the meaning, the gestures of the dramatis
personae plainly show. The mistress is interrupted,
like the maid, in the midst of her household duties ;
aud with something of terror, and something of re-
monstrance she listens to the bidding of Hermes,
whose downward-pointed staff marks the character of
his message, as the opened argumentative left hand
indicates the delivery of it. [For the gestures of
both Hermes and the woman ef. those of the same
characters on a polychrome lekythos in Stackelb.
Graber. pl. xlvii.] The Hermes may be compared
with the figure of the same god on a mirror (Etruscan)
also in Berlin, where he is superintending the removal
of a dead warrior, aided curiously enough by a youth-
ful satyr.
An interesting feature in the Berlin vase, a fifth
century piece of work, is the evident intention to
one emotion in the face of the unfortunate
ady.
very middle of the floor, and the pedestal to which
it had been attached had disappeared. It is however
evident from the discovery that at the latest in the
fourth century B.c. there existed in Cyprus a very
simple form of calathos-capital, scarcely to be derived
from the Corinthian, but belonging rather to an older
architectural system. There ave indeed species of
Cy priote capitals which have, on very slender grounds,
been claimed as ‘prototypes’ of the Corinthian ® ;
but to speak of a ‘prototype’ in this connection is
somewhat meaningless unless the word is used in a
new sense. That however the Corinthian order had
predecessors, and predecessors in a nearer degree than
the Ionic—to say nothing of the Egyptian campani-
form—is inevitable: though the gain of truth in
Vitruvius’ story may be of the minutest. Simpler
or variant forms of the calathos-capital were known
in Greek architecture,é and such a one is this from
Cyprus. But the Poli stone is something more than
a variant : it is, in despite of chronology, a predecessor,
a suggestion of the form invented by Callimachus ;
and as such deserves some attention. The purpose
it served stiil dominates it, the upper and lower
horizontal bands confining the fronds which threaten
otherwise to dissolve the basket-shape : so that what
is in effect a free and graceful architectural member is
lowered to the level of superstition and utility. Take
away the fetter of religious custom, let the fronds of
the calathos shake themselves free from the band
which still keeps up the likeness to a basket, and the
new theory of the capital stands complete. The
superstructure is no longer supported on the column
as on a prop, but it grows from it as the head of a
date-palm grows from its stem.
H. ARNOLD TUBBS.
CHERSONESOS CNIDIA.—With reference to my
article (C. 2. 1889, p. 422), I have further noticed in
the ’ApxaioA. Δελτίον of June 1888, p. 112, a new
fragment of a quota list, which shows that the
Cnidian Chersonesoi there are a συντέλεια. There
were however more than three cities, for the tribute of
the three contained in the fragment cannot jointly
amount to as much as a talent, whereas the tribute of
the Χεῤῥονήσιοι is three talents.
W. R. Parton.
Numismatic Chronicle. Part III. 1889.
Warwick Wroth. ‘Greek coins acquired by the
British Museum in 1888’ (with one autotype plate).
Describes in detail 52 of the most noteworthy of the
455 Greek coins acquired by the Museum. Among
the specimens, which include several coins not
described in the European portion of the British
Museum Catalogue of Greek Coins, are a bronze coin
of Mopsium in Thessaly with a reverse-type (Mopsus
and a Centaur) resembling a Parthenon metope (No.
III.) in the British Museum ; an unpublished Lamp-
sacene stater, of fine style, with obverse, head of Zeus ;
a silver coin of Caunus, already described by Mr. T.
Bent in Journal Hell. Stud. 1889, p. 85; a bronze
coin, witha Greek inscription, of Parlais in Lycaonia,
a town of which only Roman Colonial coins had
previously been known ; well-preserved silver coins
of Mallus in Cilicia; Jewish half-shekels of \ears 3
and 4.—Sir A. Cunningham. ‘Coins of the Tochari,
> As those in C. Ceccaldi Mons Ante du Chypre
pp. 43 and 44.
6 As in later work, like the Caryatid columns by
Kriton and Nikolaos, or the Herodian pillar of the
mosque E] Aksa at Jerusalem (the latter is figured
in Renan, Mission, pl. xli.).
THE CLASSICAL REVIEW. 71
Kushans, or Yue-ti.’ Deals chiefly with the history
and coinage of the Scythian conquerors of Bac-
triana. W. W.
ACQUISITIONS OF BRITISH MUSEUM.
1. The Trustees of the Museum have arranged for
the purchase of the magnificent silver treasure of
Chaource, which was offered for sale in Paris in June
1888 and is fully described and illustrated in the sale
catalogue issued on that occasion ; these descriptions
are partly based on a study by the Abbé Thédenat
and M. Héron de Villefosse in the Gazette Archéo-
logique of 1885, pp. 111, 256 and 317.
The treasure was discovered in 1883 by some
peasants working in a field near Montcornet (Aisne) :
the first piece was turned up by the plough, and this
led to the finding of the remainder, the entire set
having been originally wrapped in a piece of cloth
which had almost wholly perished. Six Roman bronze
coins were afterwards found on the spot, dating from
the reigns of Domitian, Trajan, Hadrian, Anton-
inus and Postumus: from this and other internal
evidence it is clear that the objects date from the
second century A.D.
They consist of thirty vases of silver, and six of
bronze plated with silver, comprising an almost com-
plete table service, méinisterium: there is also a
silver statuette of Fortuna (Zrésor, pl. II. No. 31),
with a base which may have belonged to it, a
statuette of a squatting Arab slave (7vrésor, pl. II.
No. 32) whose hair is pierced with six holes,—
possibly a pepper pot, piperatorium (?)—, and the
fragments of a flute (?). One of the silver pilates is
decorated on the interior with a medallion on which,
in relief, is Hermes between a cock and a ram (77ésor,
pl. 1X. No. 29). Perhaps the most attractive piece is
a large silver situla with movable handle (7'resor, pl.
X. No. 30), which has a broad frieze of floral orna-
ment in relief around the mouth. The reliefs in
both these pieces are parcel gilt.
The workmanship, which is of great beauty, is
probably of Roman origin, as is shown by the occur-
rence of such names as Genialis, Aurelianus, Kavari-
anus, incised on some of the objects.
2. A sard scaraboid mounted on a silver hoop and
engraved with the figure of Athene holding out the
akrostolion of a ship, found in Cyprus. Mr. Murray,
who has kindly promised to write a description of
this gem for the next number of the Review, thinks
that it may have been intended to commemorate a
naval victory in which Athene had assisted the
Cypriotes as she frequently did in the early part of
the fourth century B.c. : and he compares the Cypri-
ote silver coin (Kekulé, Balustrade d. Ath. Nike,
vignette. Six in Num. Chron. 1883 p. 334, no. 42)
on which Athene is seen seated on the prow of a ship
and holding out an akrostolion.
3. a. Bronze relief which has decorated the lower
part of the handle of a large hydria, together with
portions of the hydria to which it has belonged.
The vase when complete must have been very similar
to one which is now in the British Museum, and
which was obtained from the island of Chalke near
Rhodes ; the reliefs in both cases represent Dionysos
and Ariadne, but the one now acquired is immeasur-
ably the finer of the two; it is probably one of the
finest specimens of this class of vase decoration that
has come down to us, and may be assigned to a date
towards the middle of the third century B.c.
ὃ. A small silver cup, found with the bronze vase.
4. A series of objects from Cyprus, viz.
a. A steatite gem in the form of a negro’s head
engraved on the base with the figure of an eagle:
mounted in gold.
b. A steatite scarab: on the base is engraved a
Cretan goat.
c. An ivory tessera for the theatre, inscribed
XIII.
1" CEcIL SMITH.
SUMMARIES
The Athenaeum, 1889.
Oct. 26. Review of Leaf’s Zliad, Vol. 11. ; Notices
of Monro’s Iliad, Vol. 11. ; and of An Intermediate
Greek Lexicon. —Noy. 2. Obituary of Cobet.—
Nov. 9. Letter from J. P. Mahaffy on Cobet.—
Dec. 28. Obituary of Bishop Lightfoot.
The Academy, 1889.
Oct. 29. Notices of Marchant’s Andocides, Zimmer-
mann’s De Tacito Senecae Phil. Imitatore, and of
Reitzenstein’s Proclus’ Commentary on Plat. Rep.;
Review by A. H. Sayce of Brugmann’s Comparative
Grammar, 11. 1.—Nov. 2 Review by H. Richards
of Haigh’s Attic Theatre.—Nov. 9. Notices of Post-
gate’s Sermo Latinus, C. H. Russell’s Heceuba,
Dimsdale’s Liv. 22, Flagg’s Iphigenia in T., Hogue’s
Irregular Greek Verbs.—Nov. 16. Obituary by J. S.
Cotton cf Dr. Hatch.— Nov. 23. Review by H.
Bradley of Sonnenschein’s Parallel Grammar Series.
—Noy. 30. Letter from E. A. Sonnenschein on above
review.—Dec. 7. Review by F. Haverfield of Nettle-
ship’s Contributions to Latin Lexicography ; letter by
H. Bradley on above.—Dec. 28. Notice of Schrader’s
Sprachvergleichung.
OF
PERIODICALS.
Neue Jahrbucher fur Philologie ἃ. Padago-
gik. Ed. Fleckeisen u. Masius. (Leipzig, 1889.)
Hefte 2 and 3 contain: (1) A. Reichardt De Q.
Enni Annalibus, the first part of a paper dealing with
the language of Ennius under the headings (A) words
not used in later Latin, (B) words used by Ennius in
a peculiar sense, (0) peculiarities of grammatical
form. (2) E. Bussler Z%motheos von Gaza wu. Op-
pianos Kynegetikos, showing that the former writer
did not, as Haupt supposed, borrow from Oppian.
(3) A. Ludwich Zw den Iliasscholien, notes of some
MSS. in Spain collated in 1877 by A. Torstrik. (4)
Th. Biittner-Wobst Polybius ed. F. Hultsch, a notice
of the second edition of Vol. I. with especial reference
to Hultsch’s later views on the origin of the MSS.
(5) F. Giesing Rottenabstdnde ει. Groésze der Intervalle
in the phalanx and the manipular legion, a very
interesting paper suggested by Polybius XVIII. 29 sqq.
The effect of it is as follows. To the phalangite at
all times and to the legionary, when not in action,
3 ft. of space was allowed, of which the man himself
occupied 14 ft. But the legion, as it went into action,
‘extended’ and allowed each man another 14 ft., so
that he had 3 ft. clear for his sword-arm. The
quincunz formation was not maintained in action, for
12 THE CLASSICAL REVIEW.
the ‘extension’ of the maniples filled up the inter-
vals, which were left open, in the first instance, for
the retirement of the light troops. The intervals
were thus half as long as the front of the unextended
maniple. Polybius, loc. cit., is right in saying that
the intervals between Roman soldiers were twice as
great as between phalangites, but is pardonably
wrong in saying (or implying) that in battle one
Roman legionary would face two phalangites. There
were really two Romans to every three phalangites.
(6) H. Kothe Zu Thuk. VIL. 75, 4, proposing λιγέων
ἐπιθειασμῶν for ὀλίγων ἐπ. (7) A. Ε΄. Schone Zu
Caesar B. C. 1. 3. 3, proposing turbulentius comitinum
for urbs et ius com. (8) A. Ἐν Anspach, J. Lange and
E. Redslob Zu Plautus, several notes mostly on
Rudens and Mercator. (9) C. Angermann on
Grasberger’s Griech. Ortsnamen, an unfavourable
review. (10) J. Lange Zu Caesar B.G. V. 8—19 (the
second campaign in Britain), proposing that the
chapters should be placed in the order 8, 12, 13, 14,
9, 10, 11, 18, 15, 16, 17, 19. (11) A. Deuerling Zu
Cic. Pompeiana 18, proposing (partly after Hammer)
posse publicanos amissa wect. nostra wiect. rec. (12)
Th. Breiter Zu Manilius, emendations. (13) K. Koch
Zu Cic. pro Arch., proposing in 5 absens nobis for ab-
sentibus, and in 10 gregatim ciu. for grauat in cir.
(14) L. Triemel Cn. Flavius τι. sein Concordiatempel,
to the effect that the temple was dedicated A.v.c.
431. (15) K. Schrader Der Pannonische Triumph des
Tiberius, fixing the date of this triumph at Jan. 16th
A.D. 12. (16) K. Schliack Zu Cic. Off. III. 1, strik-
ing out interdum before colloquio.
Heft 4 contains : (1) K. Brandt Zur Gesch. etc. der
Ilias V11, a paper on the armour of Achilles in the
old μῆνις epic. (2) F. Walter Zu Tacitus, proposing in
hist. I. 71 hostes metu terreret for hostes metueret, in
hist. 1V. 23 testudinem scutorwm for test. suorwm, in
ann. ΧΙ. 35 Romani: ea cupido ete. and in ann. XV.
58 late actwm for latatwm, taking late in the sense
‘ausfiihrlich.’ (3) C. Rothe Zur Homerischen Frage,
an answer to some observations of K. Brandt in N.J.
1888 p. 513. (4) A. Ludwich Oileus u. Ileus, a collec-
tion of passages showing chiefly that Aristarchus
maintained the form Oileus. (5) F, Weck Jn Soph.
Elektra, exegetical notes. (6) F. Blass Das Newe Wei-
ner Fragment des Epicharmos, a restoration of the 10
lines and criticism of Gomperz. (7) H. v. Kleist Zu
Thukydides, a few notes. (8) R. Menge Das reciproke
Verhiltniss bei Caesar, i.e. the relation expressed by
inter se and similar phrases, enumerating the various
cases of such relation and Latin modes of expressing
it. (9) Th. Matthias Zu Ciceros Reden, emendations
chiefly to De domo. (10) W. Friedrich Zu Ciceros
Topica, a long criticism of the published texts
founded on MSS. of the xth and xith centuries. not
before collated.
Revue de Philologie. Jan.—March, 1889.
A. Hauvette, La Géographie d’ Hérodote. Are we to
call Herodotus, with H. Berger, no geographer, or,
with Bergk and Forbiger, the father of geography ?
The fact is that he broke with the theories of his
predecessors, the ‘logographi,’ such as Anaximan-
der and Hecataeus. Almost all that we know of
them comes from Herodotus. His difference from
them consists mainly in his rejection of general
hypotheses not based on certain experience. The
apparent facts of astronomy, the rising and setting
of the sun, &ec., are to his naive belief real. North
winds blow the sun farther south and so cause win-
ter, and soon. Still in astronomy he is not widely
removed from Anaximander and Anaxagoras. But
in geography he differs from Hecataeus in rejecting
what was based on theoretical system as distinct
from direct evidence. Berger holds that his criti-
cisms only spring from a vague and unintelligent ac-
quaintance with Pythagoras’ doctrine of the spherical
shape of the earth and Parmenides’ of the five zones.
But Berger's quotations from Aristophanes and Plato
do not prove that these theories were widely spread
in Greece at that time. Hdt.’s view that the north is
uninhabitable, and that Phoenician circumnavigators
of Libya could not have seen the sun to their north,
rest not on the doctrine of the five zones but on real or
supposed experience. There is no ground for think-
ing that Hdt. misunderstood the Ionian theory of a
cireumambient ocean. He rightly denied that the
earth was circular with an ocean round it ; he rightly
held that the Caspian was a closed sea, not a gulf of
the Ocean or the east ; he knew nothing, it is true,
of the sea on the north of Europe, but the accounts of
it which he rejected were fabulous. He doubted (not
denied) the existence of a western ocean and river
Eridanus ; but the Eridanus was the centre of a mass
of fable, and knowledge of the west had receded with
the Persian conquest of Ionia and destruction of its
commerce. Hdt. may well be right in accepting the
story of the circumnavigation of Libya. His view
that Libya is, in relation to Asia, a peninsula like
Asia Minor though greater, is wrong but preferable to
the artificial division of the two continents by the Nile.
He rejects the triple division, Europe, Asia, Libya ;
that into Europe and Asia he sometimes follows ;
but his fundamental doctrine is ‘the earth is one.’
This is an advance. Hecataeus made the Phasis and
the Nile flow from the external Ocean. Berger
denies this, but Hdt.’s words are clear. Popular no-
tions pictured the north of Europe as occupied by the
Rhipaean mountains, source of the north wind and
of the rivers of Scythia (the Ister included), with the
happy Hyperboreans to the north. The lonians
followed these in the main. But Hdt. denies the
Hyperboreans, ignores the Rhipaean mountains, makes
the Scythian rivers rise in marshes and the Ister in
the west. In this last point Hecataeus perhaps
preceded him. The Rhipaean Mountains were not
identified with the Alps or the Hyperboreans with
the Celts until much later dates. Hdt. does not say
that the Ister rises in the Pyrenees, but among the
Celts near a town Pyrene. The Celts probably did
not then occupy the Pyrenees, but did live round the
sources of the Danube, according to Miillenhoff. The
name Pyrene may simply be due to confusion, or may
have been applied to a place near the source of the
Danube and transferred to the Pyrenees. The false
idea that Eastern Asia Minor was a sort of isthmus
came from the Ionians. As to the course of the Nile
and the Persian gulf, Hdt. is wrong but not more so
than the Ionians. His estimates of site and distance
are only approximate and no doubt often textually
corrapt ; but they are often correct. He rejects the
traditional view that Delphi is the centre of the
earth. His views of natural phenomena, though
often mistaken, show a scientific spirit. He repre-
sents a reaction from the Ionian theories based on an
appeal to experience, which cleared the ground for
truer views.—L. Havet in Plawtus Poen. 1415 reads et
minores for eimmores.—M. Bonnet on Seneca De
Remediis Fortwitorum (fragments) agrees that the
Salmasianus is the sole authority and discusses the
text of several passages.—O. Riemann in Jac. Ann.
iv. 40 reads te invito ad te for te invite.—J. B.
Mispoulet, Le Yurbot (Juv. Sat. iv.). The assem-
bly described by Juvenal is the Imperial Council, not
the Senate. The 10 members named can all be iden-
tified, certainly or probably, with known personages,
eight of them consulars, most of considerable stand-
ing, two equites. These two, Fuscus certainly and
THE CLASSICAL REVIEW. 73
Crispinus probably, were praefectt practorio. The
story of the turbot is possibly true, but more likely
invented. The procedure &c. is no doubt true to life.
It is the only existing description of proceedings of
the Imperial Council before Hadrian reorganised
it.—H. Weil, Σύμπτυκτοι ἀνάπαιστοι. The verses so
called by Pherecrates are not, with Spiro, to be scan-
ned as Ionics a minori, but are a form of dimeter ana-
en (ree lata) ee
paestic (thus—&vdpes πρόσσχετε τὸν νοῦν) to which
the name ‘folded’ is very appropriate. These lines
are thus quite different in metrical analysis from the
catalectic glyconics which the metrical writers
wrongly identified with them and called Phere-
cratic.—E. Thomas, The Causes of Ovid's Exile.
Boissier is right in his account of the growing irrita-
tion of Augustus at the tone of Ovid’s poetry. He
and Munro hold that the immediate occasion of
Ovid’s exile was his presence at some scandalous
scene between the younger Julia and Silanus. But
if this were so, he could not have used the phrase
neptesque pias, Pont. 2, 75. A less public scandal
was quite enough to lead to the exile, 6.9. a con-
nection with the freaks or vices of one of the young
princes of the Imperial House. —R. Cagnat shows by
an examination of a large number of inscriptions that
the Roman masons must have possessed manuals
with various formulae for use on sepulchral and other
memorials ; he thereby explains many blunders in
inscriptions and quotes verses that recur with varia-
tions in many parts of the Empire.—L. Havet emends
Plautus Persa 181 and Afranius fr. 26—P. Tannery
discusses several chronological, astronomical and
mathematical texts with great acuteness.—C. Thiau-
court on Tacitus, Agricola, 45. Tac. does not mean
that he personally had anything to do with the con-
demnation and execution of Helvidius, &c. He pro-
bably was governing a province at the time. Gan-
trelle is wrong in thinking legimus, in Agr. 2, corrupt.
Tacitus’ self-accusation by implying that the whole
world is guilty is directed against the party who
cried out with excessive violence against the late
régime and its adherents. —T. Reinach on Verg. Aen.
vi. 601 sgqg. adopts in the main L. Havet’s theory of
ὙΠ τη but thinks saxwm ingens volvunt alit
escriptive of the punishment of Pirithous, not Sisy-
phus. Virgil may have broken with tradition in this
case as well as in that of Phlegyas. If so, Sisyphus
is not present at all and there is no need to suppose
a painted or sculptured original. Ixion alone keeps
his traditional punishment, because in Georg. iii.
38., iv. 484 Virgil has pledged himself to it.
There is no inconsistency, as Havet thinks, about
Theseus. He returned from the underworld in his
life but was punished there after death.—J. Baillet,
The Paean of Menchieh to Apollo and Aesculapius,
discusses the metre, and points out analogies in
Aeschylus’ Humenides ad tin. and in the paean of
the Asclepieion at Athens.—O. Riemann points out
that, in Cicero’s speeches, tantus tumque pracclarus,
tam insiynis tamque atrox, are the forms of speech
used, an order of words like tantus tam praeclarusque,
tam insignis tam atroxque not being found. Tot tam
gravesque occurs once.—O. Riemannand E. Boutroux
discuss Cicero’s argument, in De Fin. 1. 19—21 and
De Fato, 46 against the Epicurean doctrine of the
‘smallest possible’ deviation of the atoms.
Rivista di Filologia e d’Instruzione Classica.
Edd. Comparetti, Miiller and Flechia. Torino.
1889.
Fase. 10-12 contain (1) 8S. Mariotti, Plautinuwm,
a theory that the character of an ᾿Αλαζών, or Miles
Gloriosus, was suggested by a σκόλιον of Hybrias
Cretensis, Bergk pp. 1024, 1025. (2) C. Pascal,
NV te Tibulliane, three in number, the first suggesting
t t Tibullus wrote poems—perhaps the lost third
book—to Glycera (Hor. Carm. I. 33), and that the
poem IV. 13 was addressed to her; the second col-
lecting from grammarians a few citations from lost
poems ; the third proposing in III. 5, 47 an emenda-
tion, rutilis incendia classis, rutilis being gen. sing.
fem. (8) A. Pasdera, Le Oriyine dei Canti Popolart
Latini Cristiani, a discussion of the influence of
Sibylline and Jewish psalimody on early Christian
hymns. (4) G. Suster, Nuovi Emendamenti al
Panegirico di Plinio, 26 pages of emendations.
(5) C. Christofolini, Schedulae Criticae, suggesting in
Soph. Phil. 661 εἰ δὲ μὴ πάρεστ᾽ and in Antig. 4
ἀνήκοον for ἄτης ἄτερ. (6) F. Zambaldi, Un Voca-
bolario geografico di P. Bembo, an account of some
MS. geographical notes by Bembo, in the Orsini col-
lection, Codd. Gr. Vat. 1347.
Fasc. 1-3 (July-Sept. 1889) contains (1) F. Caccia-
lanza, Cecilio di Calatita e lV Ellenismo a Roma, temp.
Aug., an exhaustive collection of all the evidence
relating tu the life and works of Caecilius Calactinus.
(2) G. Suster, De Plinio Ciceronis imitatore, a com-
parison of the Panegyriews with the Pro Marcelio ;
(3) G. Fracearoli, Di aleuni luoghi controversi di
Pindaro, translations of various passages with dis-
cussion ; (4) C. Giambelli, Sugli Studi Aristotelici di
Cicerone, some observations suggesting places where
Cic. might have been indebted to the περὶ φιλοσοφίας.
Sitzungsberichte der preussischen Aka-
demie der Wissenschaften, 1889, Jan.—Aug.
1V.—Address of Prof. Th. Mommsen on the cele-
bration by the Academy of the Emperor’s birthday,—
a stately and spirited exegesis of Hor. Carm. 3, 1—6
as the initial notes of a new reign.—Reports on the
progress of the great literary undertakings of the
Academy.
XI.—XIII. Prof. Wattenbach on the MS. of the
gospels in gold letters upon purple parchment, for-
merly in the collection of the Dukes of Hamilton
(now in the hands of Mr. Quaritch). A further
defence of his earlier conjecture that the MS. was a
gift to Henry VIII. not from Leo X. but perhaps from
Wolsey, into whose possession it came from the
library at Ripon Abbey, having been made under
direction of Archbishop Wilfrid. The main theme
is prefaced by some valuable remarks on the character
and age of writing in majuscules.
XIV.—XY. Professor Kohler gives the text (with
comments) of a fragmentary inscription found in
Athens, being an account of the moneys expended by
the municipal commission of a certain year in the
purchase of gold and ivory to be used in the construc-
tion of the statue of Athena Parthenos.
XX.—XXI. By Dr. Puchstein, the conclusion of
a study of the arrangement of the figures of the gods
on the frieze of the great altar at Pergamon. The
first article was published in the Berichte for 1888
(pp. 1231 ff.). Poseidon was placed on the N. side of
the altar close to the N.W. corner. Oceanus stood
next to Doris at the end of the array of the gods at
the left side of the staircase. The place of the Gor-
gons and Erinyes, and probably of the Graeae, was
on the N. side between Poseidon and Nyx. The
Moerae with Demeter found their position on the FE.
side immediately S. of Hera and N. of Apollo.
Hermes fought at the W. end of the S. side of the
staircase. ‘The account of Apollodorus is of great
significance in the determination of these positions.
The four as yet undetermined figures on the S. side
are Theia, Hemera, Aether, and Uranus (nut Boreas).
The identification of certain of the giants follows.
γ4 THE CLASSICAL REVIEW.
XXIII.—XXIV. By Dr. Cichorius. Five Greek
inscriptions of Asia Minor from the 3rd century B.c.
to the 3rd A.D., with comments.
XXXIV. Address of Prof. Curtius in commemora-
tion of Leibniz, touching the spirit of the investiga-
tion of Greek history.
Litterarisches Centralblatt, 1889.
No. 1. Ritter, Untersuchungen iiber Plato. Die
Echtheit u. Chronologie der Plat. Schriften. ‘ Cri-
terion: language and style.’ Forty phrases are traced
through the body of Plato’s writings without decisive
results. —W. L. Newman, The Politics of Aristotle.
The text is based more on the second than on the
the first MSS. family. Introduction and notes show
a sound judgment.—Treuber, Geschichte der Lykier.
Valuable result of twenty years’ study.— Veckenstedt,
Geschichte der griech. Farbenlehre, &c. ‘The doctrine
of the development of the sense of colour cannot be
supported by philological facts.’—Wide, De sacris
Troezeniorum, Hermionensium, Epidauriorum. Col-
lection of documents, inscriptions, coins.—Herzog,
Geschichte der rim. Staatsverfassung, Vol. 11. Die
Kaiserzeit ..... bis zum Regierungsantritt Dio-
cletians. Deserves high praise.
No. 2. Arnim, Quellenstudien zu Philo v. Alex-
andria. 1. The pseudo-Philonian περὶ ἀφθαρσίας
κόσμου. 2. Philo and Aenesidemus. ‘de ebrietate ’
383-88 refer to Ae. 3. The stoic ζήτημα, εἰ μεθυσθή-
σεται ὃ copds.—Schmid, Der Atticismus in seinen
Hauptvertretern von Dionysius v. Hal. bis auf den
zweiten Philostratos, Vol. 1. Analyzes the style of
Polemo, Dio Chrysostom., Herod. Att., Lucian.
No. 3. Brugmann, Grundriss der vergl. Gramm.,
Vol. I. 1. The task of selecting from a vast mass
of material and of grouping with clearness beauti-
fully solved.—Krebs, Zui Rection der Casus in der
spdteren hist. Gracitdt, 11. “ Accusat. replaces genet.
after ἀπογινώσκειν, ἀπελπίζειν, &c. in Polybius,
Diodorus, Josephus.—Griinewald, Der freie formel-
hafte Infinitiv der Limitation im Griechischen.
‘ Phrases like ἑκὼν εἶναι have their origin in negative
clauses=‘‘as to the being willing ”’=accus. limita-
tionis.’—Birklein, Entwicklungsgeschichte des sub-
stantivierten Infinitivs. Covers the time from Homer
to Xenophon. Claims a gradual increase (Homer 1:
Thueyd. 298 : Plato 1680).—Baumgarten, Hin Rund-
gang durch die Rwinen Athens (10 Abb.). For Gym-
nasia. Due attention is everywhere paid to the re-
sults of recent investigations.
No. 4. Boltz, Hellenisch, die allgemeine Gelehrten-
sprache der Zukunft. ‘Modern Greek destined to be
the international language.’ The book contains many
happy translations from and into modern Greek. —
M. Meyer, Die Giganten u. Titanen in dev antiken
Kunst wnd Sage. ‘Giants are mythical autochthons’
(cf. Welcker J. 787). ‘Titans are traced back to old
divinities.
No. 5. Ch. Ploix, La Nature des Diewx. ‘Zeus
est le grand jour, Poseidon est le jour couvert, tous
les autre personnifient le crépuscle’ (!) —Iiener
Vorlegeblitter fiir archtolog. Ucbwngen, hrs. v. Benn-
dortf. Contains among other things a reconstruction
of Polygnot’s Iliupersis by Benndorf and Michalek.
No. 6. Deecke, Die Falisker. Deserves commen-
dation as an introduction to old-italie philology. ~
Brandt, Corpusculum poesis graec. ludibundae, tase. I.
parador. epic. graec. et Archestrati rell. It is to be
wished that B. would undertake in the same excellent
way a collection of iambic and lyric parodies,
No. 7. Fabricius, Theophanes v. Mytilene und
Quintus Dellius als Quellen der Geographie des Strabon.
A good contribution to the question of Strabo’s
sources. — Oikovouldns, of Δημοσθένους φιλιππικοί.
Of no scientific value.x—Lander, Carminis saliaris
reliquiae. Shows great learning and sound judgment.
—Hermann’s Lehrbuch der griech. Antiquitaten, 1.
Staatsaltertiimer 1. bes. von Thumser. The book
ought to have been entirely rewritten.
No. 9. Bartholomae, Beitrdge zur Flexionslehre
der indo-germ. Sprachen. The fourth essay is a dis-
cussion of the nasal in the declension of adjectives
and participles in -zé which B. claims is not original.
—Tiimpel, Die Aithiopenlinder des Andromeda-
mythos. The author’s view that Rhodes is the seat ot
the Perseus-myth is hardly to be doubted.
No. 10. Roemer, Studien zu der handschriftlichen
Ueberlicferung des Aeschylus. The Homeric ionic-epic
element in Aeschylus is much greater than in Sopho-
cles or Euripides. Many repetitions are due to his
naiveté.,—Dinarchi orationes. Adj. Demadis q. f.
νου. ὑπὲρ τ. Swdexaetias it. ed. Blass. Follows
chiefly cod. Oxon. Hiatus is avoided. Many new
conjectures are offered.
No. 11. Pichler, Virwnum (Graz) m. Atlas.
Description of the excavations near Klagenfurt
during the years 1881-83. History of Virunum.
Catalogue of all the discovered antiquities.
No. 12. Zeller, Die Philosophie der Griechen.
1, Th. 1 Abth. ‘Greatly enlarged edition.’—
Sterrett, An LEpigraphical Journey in Asia Minor
(Papers of Am. School at Athens). Situation of
Tavium ascertained through a mile-stone (time of
Nerva) ; corrected edition of the oracle published in
Bull. de corresp. hell. 1884, p. 469. The three
riddles copied by Dr. Diamantides are found—
No. 243 in Anth. Pal. xiv. 58 and 35, No. 245,
ibid. 5.—Georges, Lexicon der latein. Wortformen.
Supplement to G.’s larger dictionary. The inscrip-
tions are somewhat neglected. —Schroder, Griech.
Gotter ει. Heroen, 1. Aphrodite, Eros, Hephaestos.
S. compares Aphrodite and her nymphs with the
Sanserit Apsaras. Less probable is his view that
Hephaestos corresponds to a Sanscrit Gandharva.
No. 14. Helbig, Das homerische Epos aus den
Denkmédlern erléutert, 2 ed. Shows an increase of
100 pp. and 43 illustrations. Studniczka’s ‘ Beitrage
zur altgriech. Tracht’ have caused great changes in
the chapters dealing with dress.—Gruppe, Die griech-
ischen Culte und Mythen in thren Bexichungen zu den
orientalischen Religionen, Vol. 1. Einleitung. G. calls
his theory ‘ pure adaptionism.’ ‘ Religion is a social
instinct. This instinct is passive ; there exists only
a certain susceptibility for religious ideas, which are
created by one founder for his own interest and
diffused by missions.” The material in proof of this
theory will follow in Vel. II.
No. 15. @Q. Horatiuws Flacecus. Briefe. οὐκ]. v.
Kiessling. Of the highest importance for the exegesis
of Horace.
No. 17. H. Kluge, Zur Enstehungsgeschichte der
Ilias. (a). The pre-Homeric verse. ‘The hexameter
is the product of two trochaic tripodies, sometimes
without the last thesis of the first half-verse and
anacrusis of the second half-verse, which was allowed
to have two syllables, while all other theses consisted
of but one syllable. Along with this there existed a
tripartite verse (two dipodies and one dipody). The
Iliad, part of the hymns, and Hesiod’s poems were
composed originally without regard to quantity but
only to rhyme. Their present form is the work of
rhapsodes.’ (ὦ). Origin of the Zliad. According to
his doctrine of the original hexameter, K. considers
516 lines containing the fewest dactyls as the most
ancient lay (ujvis).—Juli Valeri Alexandri Polemi
res gest. Alexandr. Maced. transl. ex Aesopo Gracco,
ed. Kiibler. ‘The book was composed between 270
and 330.’ The text has been improved through col-
THE CLASSICAL REVIEW. τῷ
lations ot a Turin MS., which unfortunately has been
spoiled by Peyron’s chemicals.
No. 18. Λεξικὸν ἁπάντων τῶν ῥημάτων τῆς ἀττικῆς
πεζογραφικοῦ διαλέκτου ὑπὸ Ζηκίδου. Much more
copious than Veitch’s book. The first occurrence of
each verb is noted.—Kern, De Orphei Epimenidis
Pherecydis theogoniis quaestiones crit. ‘The work was
composed at the beginning of the 6th century B.c.
Epimenides’ Theogony may not have been composed
later than this time, since it was perhaps used by
Anaxagoras.’
No. 19. Frontini strategematon lib. IV. ed.
Gundermann. The text much improved by the use
of a better family of MSS. which were neglected by
fornier editors.
No. 20. Aeschylos’ Orestie mit οὐκ]. Anm. v.
Wecklein. W. proves that the text given here is
substantially the original, that many doubtful pas-
sages are genuine, and that in others conjectures are
needful. The edition makes it no longer necessary
to speak of the half-ruined condition of this master-
piece of Aeschylus.—Denkmdler griech. und rom.
Sculptur in histor. Anordnung. Unter Anleitung v.
Brunn hrsg. v. Bruckmann. A standard work.
No. 21. <Ausfiihrliches Lexikon d. griech. u. rém.
Mythologie. .... hsg. v. Roscher. Lief. 11-13.
Proves to be a most valuable book. The author of
‘Hamadryaden’ ought not to have overlooked
Mannhardt.
No. 22. H. Miiller, Das Verhdltniss des Neu-
Griechischen zu den romanischen Sprachen. Worthless.
—T. Livi ab urbe cond, libb. apparatu critico adjecto,
ed. Luchs. Vol. III]. An excellent edition.
No. 23. Mahaffv, Greek Life and Thought from the
Age of Alexander to the Roman Conquest. Ingenious,
but not always to be trusted.—ZogoxaAéous τραγῳδίαι
exd. μετὰ σχολίων ὑπὸ Γ΄. Μιστριώτου. Αἴας. No new
emendations. The commentary contains but a few
suggestions. —J. and Th. Baunack, Studien auf dem
Gebiete des griechischen wu. der arischen Sprachen.
I. Th. 2. Besides an Epidaurian inscription with
commentary the studies contain etymological analecta,
some of which are very bold, while others are happy
(as the explanation of spiritus asper in ἵππος -- ὃ ἴππος,
cf. airés=6 αὐτός, Greek negative pref. ve- = ne- in
véBpos, ‘not eating,’ cf. Lat. nefrens).
No. 24. ZL. Annaei Senecae oratorum et rhetorwm
sententiae divisiones colores, ed. H. Miiller. Besides
cod. Antverp. and Bruxell. (A, B), a Vaticanus (V)
discovered by Studemund and a second cod. Bruxell.
(D) going back to another Vaticanus (v.) have been
used for construction of archetypes. Something will
be gained by further thorough investigation into the
peculiarities of Seneca’s style, made possible by M.'s
excellent edition.—Wendorf, Lrklérung aller Mytho-
logie aus der Annahme der Erringung des Sprach-
vermogens. Worthless.
No. 25. Hygini Gromatici lib. de munitionibus
castrorum. Hrsg. u. erkl. v. Domaszewsky (3 tab.).
Cod. Arcerianus again collated and a few points of
Gemoll’s variae lectiones corrected. Some very happy
conjectures. —Bilfinger, Der biirgerliche Tag. Unter-
suchungen tuber den Beginn des Kalendartages im
class. Alterthume, &c. Contrary to the common
view and the testimonies of the ancients, that the
Greek day began at sun-set and the Roman day at
midnight, B. tries to prove that in both cases the day
began with sunrise. A detailed refutation will soon
appear in the ‘ Philologus.’
No. 26. Thumb, Untersuchungen iiber den Spiritus
asper im Griechischen. ‘ About the end of the fourth
and beginning of the fifth century every initial vowel
was pronounced with the smooth breathing, which, as
being necessary for pronunciation of the vowel is
left unmarked in most languages. There is from the
beginning a tendency to restrict the use of the spiritus
asper.—Menge u. Preuss, Lexicon Caesarianum.
Fasc. IV. Justas satisfactory as the preceding
fasciculi.
No. 27. Selections from Polybius ed. by Strachan-
Davidson (3 maps). Text based on Hultsch’s edition
with some exceptions. The selection is not always
happy.—Huemer, Die Genesis des Entschlusses in den
Tragodien des Ewripides u. Sophocles. Contains many
ingenious observations. — Merguet, Lexicon zu den
Scriften Cicero’s. 2 Th. Lexicon zu den philosoph-
ischen Schriften (1-4 Heft.). Deserves the highest
commendation.
No. 28. Polybii historiae, rec. apparat. crit. instr.
ΕΒ, Hultsch, Vol. I. ed. alt. The text is based on
cod. Vatican., the authority of which seems over-
estimated in some cases. The preface is re-written
and not a few passages of the text are corrected.—
Meusel, Lexicon Cacsarianum, Fase. IX.-XIII. A
standard work.—Herzog, Studien zwr Geschichte der
griech. Kunst (6 plates). Somewhat superficial. The
first part deals with groups ; the second part with the
series of gods on vases of lower Italy. The plates
contain some inedita.
No. 29. Lezius, De Alexandri Maced. exped. indica
quaestiones. ‘ Arrian, based on Ptolemy and Aristo-
bulos, forms the foundation for the history of
Alexander inclusive of his Indian expedition. Dio-
dorus, Curtius and Justin are based on popular
traditions and can supplement Arrian’s report in some
particulars.’—Handbuch der classischen Alterthwms-
wissenschaft, hrg. v. Iwan Miller, Vol. VII.—Christ,
Griech. Litteraturgeschichte. A handy and useful
book, although many details are open to dispute.
No. 30. Hesselbarth, Historisch-kritische Unter-
suchungen zur dritten Dekade des Livius. Although
in general there ought to be more exactness and clear-
ness, the book contains not a few valuable suggestions,
—Ciceronis pro M, Caecilio oratio, ad opt. codd. denuo
coll. recogn. Vollgraf, accedit apparat. critic. (Leyden),
Cod. Parisinus and Salisburgensis collated. Text
shows few improvements. As pupil of Cobet, V. is
apt to undervalue the MSS. authority.
No. 31. Rossbach, De Senecae philos. libb. recen-
sione et emendatione. Insunt Senecae frgg. palatina
ed. a G. Studemund. Continuation of R.’s ‘De
Senecae filii scriptis’ and ‘De Senecae dialogis.’
‘For the text of the dialogues a cod. Par. saec. xiii.
should not be neglected; for the consolatio ad
Polybium, a group to which a Vatic., Paris saec. Xiv.,
and Urb. saec. xv. belong; for de clementia, besides
cod. Naz., two Paris. saec. xiii. going back to a widely
differing archetype; for the epistul. especially a
Florent.’ Then follow a classification of younger
MSS., an enumeration of about 20 Vatic. and Paris.
codd. saec. xii.-xiii. not yet collated, orthographica,
emendations, Appendix on Florus’ Epitoma de T.
Livio, which is thought to be an epitome of the
histories of Seneca rhet.
No. 32. Immerwahr, Die Lakonika des Pausanias.
The main sources of the 3rd book of Pausanias are
χρόνων avaypaph (especially used for ch. 1-10, 5) and
περὶ τῶν ἐν Λακεδαίμονι θυσιῶν (for ch. 10, 6-21, 3),
written by Sosibios of Lacedaemon. Besides these,
for the histor. introduction Herodotus, Ephorus,
Theopompus, Phylarchus are used ; for the periegesis
of Amyclae, Polemon. The source of 21, 4-26, 11
seems to be a geographical manual of the first century
after Christ.—Th. Klette, Leonardi Arctini ad Petrum
Paulum Astrwm dialogus. Zum ersten Male.....hrsg.
No. 33. Johs. Seger, Byzantinische Historiker des
x. u. xi. Jhd. 1. Nicephorus Bryennios. The family
of the Bryennii, sources, style, contents of Nice-
76 THE CLASSICAL REVIEW.
phoros’s Chronicle.—Danielsson, Grammatische wnd
etymologische Studien, 1. κάρα, κέρας and its family.
—Rénsch, Semasiologische Beitrage zum lat. Worter-
buche 11. Adject. u. Pronom., Adverb u. Adverbialia.
Collection of rare words and words with uncommon
meanings, mostly gathered from the works of eccle-
siastical authors. —Le Bas, Voyage archéologique
(1842-1844)..... publ. et comment. par S. Reinach.
While the inscriptions of the original work have been
left aside entirely, the Itineraire, Monuments figures,
and Architecture are given in full, and the number of
tables is considerably increased. The text is re-
written for the most part by Reinach.
No. 34. Noni Marcelli compend. doctr. em. adn.
Luc. Mueller, 1. Shows all the peculiarities of its
author : it has brilliant conjectures, but little atten-
tion is paid to minute questions. The most important
MSS. have been compared (not always satisfactorily,
as M. himself admits).—Jorgensen, Kvindfiguren ὁ
den archaiske graeske kunst. (med 20 Afb.) adject.
est argum. latine conscript. A catalogue of the
archaic statues of women found lately on the Akropolis
of Athens and similar statues of Eleusis, Delos and
Boeotia. The second part deals with the difference
between Dorian and Ionian costume.
No. 36. Tatiani oratio ad Graecos rec. Schwartz.
Cod. Paris. 451 collated; many new conjectures ;
excellent indices. Type and paper poor.—AioxvAov
Προμηθεὺς δεσμώτης ἔμμετρος mapapacis μετ᾽ εἰσαγωγῆ 5
καὶ σημειώσεων ὑπὸ Ἐανθοπούλου. Introduction con-
tains nothing new. Text based on Wecklein’s ed.
‘The paraphrase does not give an idea of the grandeur
of the original.—Aristotelis quae feruntur de plantis,
de mirab. auscult., mechanica, de lin. insec., ven-
torum sit. et nom., de Melisso Xenophane Creorgia
ed. Apelt. The text of ‘de lin. insec.’ and ‘de
Melisso’ is renovated and greatly improved. The
first is based on Vatic. 258 without neglecting
Laurent. 87. 21. The MSS. of the second are divided
into two classes: I. class represented by cod. Lip-
siensis, on which the text is based ; II. class repre-
sented by Vatic. 1302, all going back to a sometimes
illegible archetype. The text of the other writings
shows little improvement.
No. 37. Ohnesorge, Die rémische Provinzliste von
297. I.—An excellent discussion of this important
list.—Apostolis, Lettres inédites, publ. daprés les
MSS. du Vatican, p. H. Noiret.—Very good edition
of A.’s letters. Important for the history of Philology
and the Renaissance.
No. 38. Topffer, Attische Genealogie. This dis-
eussion of the (mythical) origin of the families of the
Attic nobility is a valuable contribution to Attic his-
tory and archaeolory.—Welzhofer, Geschichte des
griechischen Volkes bis...Solon. Written in an attrac-
tive style. W. shows an antipathy against modern
criticism of the ancient traditions. —T7ragicorum
Graecorum fragmenta, rec. Nauck, ed. 11. The num-
ber of fragments is increased (especially from Egyp-
tian papyri), the text often emended, the results of
recent investigation are carefully given. (Half a
column of the reviewer's (H. St.) conjectures fol-
lows).
No. 39. Hiller, Beitrdge zur Textgeschichte der gricch-
ischen Bukoliker.—An excellent investigation. It is
the first attempt—after Ahrens in Philol. xxx.—to
classify the MSS., and is at the same time a critical
edition of 10 poems, with apparatus and prolegomena.
An invaluable basis for a future critical edition of the
Bucolic poets, especially Theocritus.—Josephi opera
omnia post ..Bekkerum rec. Naber I.—Based on
Niese’s (‘fortasse non semper accuratissimi’) colla-
tions, but independent in textual matters, often
deviating from WNiese’s extremely conservative
edition.—Lacher, Die Handschriften und Classen der
Aristophanesscholien.—‘The scholia go back to an
archetype saec. x., of which abstracts and copies were
made. In the xii. century Tzetses and the author of
the Byzantine commentary in M, in the xiii. and xiv
among others Triclinius wrote new commentaries, It
will be the task of a future editor to construe the
original archetype of the scholia.’—Bilfinger, Die
antiken Stundenangaben.—Claims that almost with-
out exception hora sexta means ‘at 12 o'clock,’ not
‘the time from 11-12.’
No. 40. Duruy, Geschichte des rémischen Kaiser-
reichs...iibers. v. Herzberg. vol. IVY and V. As
satisfactory as the preceding volumes.—Jamblichi
Protrepticus ad fid. cod. florentini ed. Pistelli.—An
excellent edition of the Protrepticus, which in some
parts has the value of a MS. for Platonic writings.
Vitelli has contributed a number of conjectures. A
good index nominum and index verborum are added.
No. 41. Koetschau, Die Textiiberlicferung der
Biicher des Origenes gegen Celsus. Excellent prolego-
mena to a critical edition of Orig. contr. Cels.—
Schreiber, Die hellenistischen Relicfbilder erlautert,
I. Contains ten very good reproductions. The text
will follow later.—Baumeister, Bilderhefte aus dem
griechischen und rimischen Alterthum fiir Schiiler.
Contains selections from B.’s ‘Denkmaler’ and de-
serves a wide circulation.
No. 43. Diinzelmann, Der Schauplatz der Varus-
schlacht. A noteworthy, short essay. Considers the
Λούπιας of Strabo 7, 1, 5 (—the emendations of
Bergk in Rh. Mus. 1882, 297 are unknown to D.—)
as the modern Hunte, Aliso then being Hunteburg,
the Venner Miihlenbach or Else the ’EAtowy of Dio.
Hard to reconcile with Dio 54, 33, is D.’s assumption
of the Bructeri dwelling between Ems and Hunte.—
Joannides, Sprechen Sie Attisch? An excellent little
book which is warmly to be recommended.—Plessis,
Traité de Métrique Grecque et Latine. The Greek
metres are somewhat neglected. It is intended for
‘étudiants de nos facultés’ and ‘professeurs des.
établissements d’enseignement secondaire,’ and will
very well answer its purpose.
No. 44. Joh. Schmidt, Die Plwralbildung der in-
doyermanischen Neutra. The main idea of the highly
interesting investigation is that the neuter plurals
were originally fem. sing. used collectively.—Ebert,
Allgemeine Geschichte der Litteratur des Mittelalters
im Abendlande bis zum Beginne des XI. “μα, Re-
vised and enlarged edition of this excellent book.
No. 45. Wieseler, Archéologische Beitrage, II.
Collection and discussion of figures supposed to be
representations of Aesculapius.
No. 46. Ribbeck, Geschichte der rimischen
Dichtung, 11. Deals with the poets of the Augustan
age. t.’s original investigations and mature judg-
ment, as well as his lively and attractive style, make
the book pleasant and profitable reading, both for
scholars and the general public.—Overbeck, Griech-
ische Kunstmythologie, 111. 5; Apollo, Lief. 1 and 2.
More concise than the treatise on Zeus. The discus-
sion of the so-called ‘canonical ideal’ is rightly
omitted, the number of illustrations increased and
improved, the enumeration of monuments is almost
complete, the text full of suggestions and valuable
results.
No. 47. Pomtow, Beitrage zur Topographie von
Delphi. Will serve as a good basis for future inves-
tigations, since it gives the complete results of all
the preceding excavations.—Puchstein, Das ionische
Capitell, An original and very valuable investigation.
THE CLASSICAL REVIEW.
LIST OF PROGRAMMES ISSUED IN
E, Anspach.—Die Horazischen Oden des ersten
Buches in Bezug auf Interpolation, Aufbau und Zeit
der Abfassung. Ii. Gymm. zu Cleve. 40 S. [On
Metrik. ]
Rk. Bartels.—Beziehungen zu Athen und seiner
Geschichte in den Dramen des Euripides. Joachim-
thalsches Gymn. zu Berlin. 208.
α΄, Brambs.—Uber Citate und Reminiszenzen aus
Dichtern bei Lucian und einigen spateren Schrift-
stellern. Eichstatt. 83 8S.
K. Davin.—Beitrage zur Kritik der Quellen des
ersten punischen Krieges. Grossh. Gymn. zu Schwerin.
41 5. 1889. [On the Regulus-Embassy. ]
J. v. Destinon.—De Flavii losephi bello Iudaico
recensendo ad Benedictum Niese epistula critica.
symn. zu Kiel. 17 5.
Κ΄. Deutschmann.—De poesis Graecorum rhythmi-
eae usu et origine. Gymn. zu Coblenz. 29 S. [On
accent-poetry of Byzantines of fourth cent. ]
A. Dréger.—Ovid als Sprachbildner. Gymn. zu
Aurich. 8. 19 S. [392 new formations, of which 153
are a.A.]
E. Eckardt.—De temporum ratione, quae Tra-
chiniis fabulae Sophocleae subest, et de eiusdem
fabulae parodi contextu. Gymn. zu Salzwedel.
12 5.
Ehwald.—Ad historiam carminum Ovidianorum
recensionemque symbolae. Gymn. zu Gotha, 20 S.
[Mainly on the Tristia. ]
P. Esternaux.—Die Komposition von Frontins
Strategemata. Franz. Gymn. zu Berlin. 23 S.
[Defence of authenticity of Bk. iv. ]
W. Friedrich.—Varietas lectionis codicis Vossiani
LXX ad Ciceronis libros qui vulgo de inventione
vocantur duos. Gymn. zu Miihlhausen. 8°. 38S,
W. Gemoll.—Beitrige zur Kritik und Erklarung
von Xenophons Anabasis. II. Gymn. zu Kreuzburg.
33S. [Mainly on use of article].
Gropius, k.—Isidor Hispal. Etymol. XIII. 13 (de
diversitate aquarum) als Handhabe zur Beurteilung
von Isidorus- Handschriften. Gymn. Weilburg.
10 8.
W. Hahn.—Zeus in der Ilias.
Stralsund. 28S.
EH. Hasse.—Uber den Dual bei Xenophon und
Thukydides. Gymn. zu Bartenstein. 215%.
W. Heraeus, Vindiciae Livianae. 1. Gymn. zu
Hanau 168. [Comparison of Livy and Tacitus in
language].
£. Hermes.—Kritische Beitrage zu den Briefen des
sore Pat ΤΠ. Annaeus Seneca. Gymn. zu Moers.
148.
Chr. Herwig.—Das Wortspiel in Ciceros Reden.
Gymn. zu Attendorn. 19S.
H, Heubach.—Quibus vocabulis artis criticae pro-
priis usi sint Homeri scholiaste. Realgymn. zu Kise-
nach, 238.
H. Karbawn.—De origine exemplorum, quae ex
Ciceronis scriptis a Charisio, Diomede, Arusiano
Messio, Prisciano, aliis grammaticis latinis allata
sunt. Gymn. zu Wernigerode. 18S.
Il. Gymn. zu
~J
—
1889.
: D. Kennerknecht.—Zur Argonautensage. Bamberg
0S.
P. Kleber.—Die Rhetorik bei Herodot.
progymn. zu Lowenberg. 27 8.
P. Krumbholz.-—De Ctesia aliisque auctoribus in
Plutarchi Artaxerxis vita adhibitis. Gymn. zu Eise-
nach. 258.
Fr. Liesenbery.—Die Sprache des Ammianus Mar-
cellinus. 1. Gymn. zu Blankenburg 21 5. [On
foreign-words].
H. Linke.—Studien zur Itala.
zu Breslau. 28 S.
A. Lowinski.—Zur Kritik der Horazischen Sati-
ren. Gymn. zu Deutsch-Krone. 18 5. [On inter-
polations. |
Ff. Macke.—Die rémischen Eigennamen bei Taci-
tus. III. Gymn. zu Hadersleben. 22 S.
Georg Meyer.—Der gegenwartige Stand der Thuky-
dideischen Frage. Klosterschule zu Ilfeld. 42 8S.
[Against Herbst etc. ].
Petrus Meyer.—Quaestiones Platonicae. Gymn. zu
M. Gladbach. 25 S. [Polemical against Teichmiiller].
E. Mollmann.—Herodots Darstellung der Ge-
schichte von Cyrene. Kneiphofisches Gymn. zu
Konigsberg. 24 5.
S. v. Monsterberg-Miinckenau.—De concentu trium
Aristotelis de voluptate commentationum priorisque
Nicomacheorum fide. Konig Wilhelm-Gymn. zu
Breslau. 8. 4 5.
C. F. W. Miller.—Kritische Bemerkungen zu
I. Real-
Elisabeth-Gymn.
Plinius’ naturalis historia. Johannes-Gymn. zu
Breslau. 288. [Defence of traditional text against
Mayhoff].
Ernst Miller.—Ciceros Rede de provinciis consu-
laribus, verdeutscht. Gymn. zu Kattowitz, 1858.
H, Peters.—Beitrage zur Heilung der Uberliefer-
ung in Quintilians institutio oratoria. Realgymn.
zu Cassel. S. 16—25. [On foreign words].
Pflug.—Diodor und Livius als Quellen fiir den
zweiten Samniterkrieg. Gymn. zu Waldenburg.
16 S.
Rauch.—Gerundium und Gerundivum bei Curtius.
Gymn. zu Meiningen. 218. [Statistics].
A. Reeck.—Beitrage zur Syntax des Catull.
gymn. zu Bromberg. 18S.
Rk. Richter.—Kritische Bemerkungen zu Ciisars
Commentarius VII, de bello Gallico. Gymn. zuStar-
gard. 39S. -
Rich. Schneider.—Zwei Reden. I. Der Prome-
theus des Aschylus. II. Die Medea des Euripides.
Gymn. zu Duisburg. 9S.
E. Rich. Schultze.—Quaestiunculae grammaticae
ad oratores Atticos spectantes. Gymn. zu Bautzen.
318. [Statistics].
Thalheim.—Quaestiones Demosthenicae. Gymn.
zu Schneidemithl 13 8. [On Callistr. Mantith.
Theokr. ]
CU. Wetzell.—Lexici Antiphontei specimen. Gymn.
zu Laubach. 18S. [Down to ἀμφότεροι].
HT. Wéindel.—Demosthenis esse orationem, quae
περὶ συντάξεως inscribitur. Gymn. zu Hameln, 22 8.
Real-
78 THE CLASSICAL REVIEW.
LIST OF NEW BOOKS.
ENGLISH BOOKS.
Notes on Abbreviations in Greek
Allen (T. W.)
Royal 8vo. Sewed. Oxford, Cla-
Manuscripts.
rendon Press. 5s.
Aristotle. Nicomachean Ethies, Books 1-4 and Book
10, Cap. 6 to end, analysed, annotated, and trans-
lated for Oxford Pass-men by S. H. Jeyes, M.A.
8vo. W.H. Allen. 6s.
Baker (A.) Latin Prose for London Students. 12mo.
pp. 94. Belland Sons. 25.
Caesar. Commentaries, Book 8. With English notes
by George Long, M.A., and with Vocabulary by
W. F. R. Shilleto, M.A. Feap. Bell and Sons.
15. θα.
Gallic War, VI. pp. 91. C. Colbeck.
millan and Co. 18. 6d.
Cicero. De Amicitia and Pro Balbo. Translated by
J. Gibson. Svo. Sewed. Cornish and Son. 2s. 6d.
Pro Balbo. With Introduction, Notes, Vocabu-
laries, and Translation. Post 8vo. Clive. 2s. 6d.
Edgar (J.) Alphabetical Index to the Greek Termin-
ations. Feap. Cornish and Son. 15. 6d.
Euripides. Uphigeneia at Aulis. With Introduction
and Notes by Clinton E. S. Headlam. 12mo.
pp. 156. Cambridge Warehouse. 2s. 6d.
Harper (W. R.) and Weidner (R. F.) An Introduc-
tory New Testament Greek Method, together with
a Manual containing Text and Vocabulary of
Gospel of John and Lists of Words, and the Ele-
ments of New Testament Greek Grammar. 8vo.
pp. 520. D. Nutt. 7s. 6d.
Harvard Studies in Classical Philology, edited by a
committee of the classical instructors of Harvard
University. Vol. I. 1890. 8vo. 206 pp. Plates.
4s,
Jevons (F. B.) A History of Greek Literature from
the earliest period to the death of Demosthenes.
Second Edition. With Appendix and Examination
Questions for the use of Students. Post 8vo.
pp. 536. Griffin. 8s. 6d.
Mac-
King (J. E.) and Cookson (C.) Introduction to the
Comparative Grammar of Greek and Latin, Crown
8vo. Oxford, Clarendon Press. 5s. 6d.
Millington (H.) Translation into Latin Verse. Feap.
Vellum. Bell and Sons. 2s. 6d.
Pliny’s Letters. Books I. and II. With Introduc-
tion, Notes, and Plan. Edited by J. Cowan. 12mo.
pp. 232. Macmillan and Co. 5s.
Preston (G.) Exercises in Latin Verse of various
kinds: Key. Feap. Macmillan and Co. 5s.
Psalms in Greek, according to the Septuagint. Edited
by Henry B. Swete, D.D. Crown 8vo. Cambridge
University Press. 2s. 6d.
Sargent (J. Y.) Exemplaria Graeca: Selections from
passages for translation into Greek. 12mo. pp. 106
Frowde. 958,
Models and Materials for Greek Iambic Verse.
Crown 8vo. Clarendon Press. 45, 6d.
Sanday (W.) Appendices to the Greek Testament.
Feap. Oxford, Clarendon Press. 3s. 6d.
Tacitus. Histories, Books 3, 4, and 5. With Intro-
duction and Notes by A. D. Godley. 12mo.
pp. 804. Macmillan ἃ Co. 5s.
Theocritus, Bion, and Moschus rendered into English
Prose, with Introductory Essay by A. Lang. Large
Paper Edition. 8vo. Macmillan and Co. 95.
Thucydides. Book 1V. A Revision of the Text,
illustrating the principal causes of corruption in
the Manuscripts of this author, by William Gunion
Rutherford. 8vo. pp. Ixxviili and 134. Mac-
millan and Co. 7s. 6d.
Virgil. Aeneid, Books IX. and X. A Translation
by A. A. Irwin Nesbit. Post 8vo. pp. 50. Clive.
15. 6d.
— Georgics, Lib. I. Edited. for the use of schools,
by T. E. Page, with Vocabulary. 18mo. pp. 120. °
Macmillan and Co. 1s. 6d.
Waddell (W. W.) Versions and Imitations in Greek
and Latin. Crown 8vo. 2s. 6d.
FOREIGN BOOKS,
Aristaenctus. Lettere, tradotte da un Accademico
florent. 24mo. 98 pp. Soriano nel Cimino
Capaccini. 2 lire.
Baumeister (A.) Bilderhefte aus dem griechischen
und romischen Altertum fiir Schiiler. Heft 3.
4to. Miinchen, Oldenbourg. 1 Mk. 25.
Contents : Sagenkreis des trojanischen Krieges.
pp. 77-116.
Bénard (C.) L’Esthétique d’Aristote. ὅνο. 371 pp.
Paris, Picard.
[Extract Compte rendu des Séances et Travaux
de I’ Académie des Sciences morales et politiques].
Boetticher (E.) La Troie de Schliemann une Nécro-
pole ἃ Incinération ἃ la maniére assyro-baby-
lonienne. Avec préface parC. de Harlez. [Extract
Muséon.] 8vo. vii, 115 pp. 12 plates. Louvain.
6 Mk.
Boot (J. C. G.) De historia gymnasii Leovardiensis
ab origine ad hance aetatem. Editio altera emen-
data. 8vo. 8, 119 pp. Amsterdam, van Looy.
1 ΕἸ. 25.
Breusing (A.) Die Lésung des Trierenratsels ; die
Inrfahrten des Odysseus, nebst Erginzungen und
Berichtigungen zur Nautik der Alten. 8vo.
vii, 124 pp. Bremen, Schiinemann. 3 Mk 50.
Broccardi (L.) Grammatica latina secondo i metodi
pit recenti. Parte II]. Morfologia. 8vo. 181,
684 pp. Torino, V. Bona. θ lire.
Buresch (K.) Klaros, Untersuchungen zum Orakel-
wesen des spiteren Altertums, nebst einem Anhang
das Anecdotum χρησμοὶ τῶν ἑλληνικῶν θεῶν ent-
haltend. ὅνο. iii, 134 pp. Leipzig, Teubner.
3 Mk. 60.
Caesar. Commentariicum Supplementis A. Hirtii et
aliorum. Iterum recognovit et adnotationem
criticam praemisit E. Hoffmann. 2 vols. 12mo.
Wien, Gerold’s Sohn. 4 Mk. 50.
Calonghi (F.) Li’aoristo greco particolarmente in
Omero : contributo alla sintassi storica della lingua
ereca. ὅνο. 31 pp. Torino, V. Bona.
Capellanus (G.) Sprechen Sie Lateinisch ? Moderne
Conversation in lateinischer Sprache. 12mo. 108 pp.
Leipzig, C. A. Koch. 1 Mk. 60.
Censorini de die natali liber. Ad codicum denuo
collatorum fidem recensuit J. Cholodniak. ὅνο.
iii, 79 pp. St. Petersburg. Ricker. 1 Mk.
THE CLASSICAL REVIEW. 79
Chassang (A.) Selectae Narrationes e scriptoribus
latinis. Avee des notes des arguments et des
modeéles d’analyse littéraire, ἃ l’usage des classes de
seconde. 16mo. viii, 307 pp. Paris, Hachette.
2 fr. 25.
Cicero. Brutus. Fiir den Schulgebrauch erklirt von
K. W. Piderit. 3te Auflage besorgt von W.
Friedrich. S8vo. iv, 306 pp. Leipzig, Teubner.
2 Mk. 25.
—w— in M. Antonium oratio Philippica secunda.
Scholarum in usum edidit A. Kornitzer. 16mo.
77 pp. Wien, Gerold’s Sohn. Boards. 70 Pf.
—in C. Verremaccusationis liber LV. Scholarum
in usum edidit A. Kornitzer. 16mo. 127 pp.
Map. Wien, Gerold’s Sohn. 90 Pf.
Cornelius Nepos: Vitae Selectae. In usum scholarum,
edidit R. Bitschofsky. 16mo. viii, 106 pp. Map.
Wien, Gerold’s Sohn. Boards. 80 Pf.
Corpus glossariorum latinorum a G. Loewe inchoatum,
auspiciis societatis litterarum regiae Saxoniae com-
posuit, recensuit, edidit G. Goetz. Vol. IV. 8vo.
Leipzig, Teubner. 20 Mk.
Contents: Glossae codicum Vaticani 3321,
Sangallensis 912, Leidensis, 67. xlili, 605 pp.
Corpus inscriptionum latinarum, consilio et auctori-
tate Academiae litterarum regiae borussicae editum
Vol. Lil. supplementum. fase. 1. folio. Berlin,
G. Reimer. 17 Mk.
{[Contents: Inscriptionum Ilyrici latinarum
supplementum, ediderunt T. Mommscen, O. Hirsch-
feld, A. Domaszewski. Fasciculus 1. 1199,
1372 pp].
Dissertationes philologicae Halenses.
306 pp. Halle, Niemeyer. 8 Mk.
Draheim (H.) Sophokles. Chore. Ein Fiihrer
durch die Tragédien des Dichters, 12mo. iv, 75 pp.
Kisenach, Bacmeister. 80 Pf.
Dumon (K.) Le Théatre de Polyciéte.
Vol. X. 8vo.
Reconstruc-
tion d’aprés un Modele. Folio. 51 pp. 3 plates.
Haarlem. 16 Mk.
Fink (J.) Der Verschluss bei den Griechen und
Romern. 8vo. 58 pp. Plates. Regensburg.
Bauhof. 2 Mk.
Florilegium graecum in usum primi gymnasiorum
ordinis collectum a philologis Afranis. 4 fasciculi.
12mo. vi, 72, 71, 74, 70 pp. Leipzig, Teubner.
Each 45 Pf.
Garrer (A. H.) Schonaeus. Bijdrage tot de ge-
schiedenis der Latijinsche school te Haarlem.
8vo. 76 pp. Haarlem, F. Bohn. 1 fl.
Giri (G.) Deltradurre presso i latini. 8vo. 182 pp.
Milano, D. Briola. 2 lire 50.
Goetzeler (L.) Quaestiones in Appiani et Polybii
dicendi genus cum appendice de codicibus quibus-
dum Herodianeis. 8vo. 143 pp. Wiirzburg,
Stahel. 3 Mk. 60.
Grammatici graeci recogniti et apparatu critico in-
structi. Partis IV. volumen 1. 8vo. Leipzig,
Teubner. 14 Mk,
Contents: Theodosii Alexandrini canones, G.
Choerobosci scholia, Sophronii patriarchae Alexan-
drini excerpta, recensuit et apparatum criticum
indicesque adjecit A. Hilgard. Vol. 1. Theodosii
canones et Choerobosci scholia in canones nominales
continens. vi, 417 pp.
Groh (K.) Geschichte des ost-rémischen Kaisers
Justin 11. Nebst den Quellen. 8vo. viii, 120 pp.
Leipzig, Teubner. 3 Mk. 20.
Gurlitt (W.) Ueber Pausanias.
8vo. xii, 494 pp. Graz, 1890.
Lubensky. 10 Mk.
Gutschmid (A.) Kleine Schriften. Herausgegeben
von F. Ruhl. Band I. Schriften zur Aegypto-
logie und zur Geschichte der griechischen Chrono-
Untersuchungen.
Leuschner and
graphie. 8vo. xii, 573 pp. Portrait.
Teubner. 14 Mk.
Hartman (J. J.) Analecta Xenophontea nova.
vii, 352 pp. Leiden. 10 Mk.
HSIOAOY τὰ ἅπαντα ἐξ ἑρμηνείας K. Sittl. Vol. I.
8vo. Athens, C. Beck. 10s.
Heusser (Th.) Griechische Syntax zum neuen Testa-
ment, nebst Uebungsstiicken zum Uebersetzen ins
Griechische fiir Formenlehre und Syntax. Mit
Vorwort von F. Haarbeck. 12mo. vii, 88 pp.
Basel, Spittler. Bound. 1 Mk. 50.
Homer. Carmina, recensuit et selecta lectionis
varietate instruxit A. Ludwich. Pars 11. Odyssea.
Vol. 1. 8vo. xxviii, 315 pp. Leipzig, Teubner.
8 Mk.
De Odussee. Metrisch vertaald door C. Vos-
maer. 8vo. 344 pp. Leiden, A. W. Sijthoff.
8: 50;
Horatius recensuit atque interpretus est J. G. Orellius.
Editio IV. major emendata et aucta. Vol. II.
Satirae. Epistulae. Lexicon Horatianum. Post
J. G. Baiterum curavit W. Mewes. Fasciculus 1.
Leipzig,
8vo.
8vo. ii, 160 pp. Berlin, Calvary and Co.
Each 3 Mk.
Ihm (G.) Materialen fiir die Ovidlektiire. ὅνο.
xli, 60 pp. Paderborn, Schoningh. 1 Mk.
Ihne (W.) Réomische Geschichte. Band VII. Die
Birgerkriege bis zum Triumvirat. 8vo. iii, 483 pp.
Leipzig, Engelmann. 5 Mk.
Jahrbuch des kaiserlichen deutschen archaeologischen
Instituts. 2s Erganzungsheft. 4to. Berlin,
G. Reimer. Boards. 24 Mk.
Contents: Altertitimer von Aegae unter Mit-
wirkung von C. Schuchhardt herausgegeben von
R. Bohn. 68 pp. 75 ents.
Ian (C.) Vorlagen zu Uebungen im lateinischen Stil
fiir Sekunda. 8vo. vi, 60 pp. Leipzig, Teubner.
lowed.
Kaegi (A.) Griechische Schulgrammatik. Mit
Repetitionstabellen als Anhang. 2t¢ vielfach ver-
tinderte und verbesserte Auflage. 8vo. xviii.
286, xlvi pp. JBerlin, Weidmann. Bound.
3 Mk
Kekulé (R.) Ueber die Bronzestatue des sogenannten
Idolino. 49. Programm zum Winckelmannsfeste
der archiaologischen Gesellschaft zu Berlin.
4 plates. 4to. 22 pp. Berlin, G. Reimer.
2 Mk. 80.
Kovtoyovns (3. A.) Διορθώτικα εἰς τὰ ἔθνικα
Στεφανοῖ τοῦ Βυζαντίου ἐν οἷς ἀκριβὴς περὶ τοῦ
Σιμίας λόγος: Ἑλληνικόν ὀνομάστικον περιλαμβάνον
τὸ πρῶτον ἐξ ἐπιγραφῶν γνωριζόμενα ὀνόματα.
Ἑκατοντὰς α΄. gr. 8vo. 97 pp. Jena, Doebereiner.
Képke (R.) Die lyrischen Vermasse des Horaz. Fir
Primaner erkliirt. 416 Auflage. 8vo. 32 pp.
Berlin, Weidmann. 60 Pf.
Kronenberg (A. J.) Minuciana sive annotationes
criticae in Minucii Felicis Octavium. Specimen
litterarium inaugurale. 8vo. v, 90 pp. Leiden,
van Doesburgh. 2 Mk.
Labbé (E.) Syntaxe latine nouvelle, abrégée pour
iUsage des Classes. 12mo. xvi, 126 pp. Paris,
Thorin.
(Laloux et Monceaux). Restauration d’Olympe.
L’Histoire les Monuments, le Culte, et les Fétes.
Folio. 232 pp. 20 plates and illustrations. Paris,
Quantin. 100 fr.
Lindenschmidt (L.) Das rémisch - germanische
Central-Museum in bildlichen Darstellungen aus
seinen Sammlungen. 4to. 11 pp. 50 plates.
Mainz, von Zabern. Boards. 15 Mk,
Livy. Liber XXII, fiir den Schulgebrauch erklart
von K. Tiicking. 8:8 verbesserte Auflage. 8vo.
121 pp. Paderborn, F. Schoningh. 1 Mk. 20,
80 THE CLASSICAL REVIEW.
Livy. Apparatu critico adjecto edidit A. Luchs.
Vol. 1V., libros XXVI.-XXX. continens, 8vo.
xi, 295 pp. Berlin, Weidmann. 3 Mk.
recognovit J. Sommerbrodt. Vol. I.
Tawianis,
pars 2. 8vo.
6 Mk.
Mahly (i.)
Sophokleisches.
20.
Mahn (A.) De Dionis Chrysostomi codicibus.
90 pp. Leipzig. 2 Mk.
Marcelli, de medicamentis liber.
reich. 12mo. vi, 414 pp.
3 Mk. 60.
Menge (R.) et S. Preuss. Lexicon Caesarianum.
Fasciculi IX. et X. 8vo. Columns 1025-1280.
Leipzig, Teubner. Each 1 Mk. 60.
Merguet (H.) Lexikon zu den Schriften Cicero’s mit
Angabe siimtlicher Stellen. Theil 11. Lexikon zu
den philosophischen Schriften. Heft 5 und 6.
(Lieferung 17-24). 8vo. 641, 998. Jena, Fischer.
15 Mk.
Miiller (G. A.) Christus bei Josephus Flavius. Eine
kritische Untersuchung als Beitrag zur Losung der
beriihmten Frage und zur Erforschung der Ur-
geschichte des Christentums. 8vo. ili, 53 pp.
Innsbruck, Wagner. 1 Mk. 60.
Neudecker (G.) Der klassische Unterricht und die
Erziehung zu wissenschaftlichem Denken. Eine
kritische Untersuchung. 12mo. 39 pp. Wurzburg,
Stuber. 60 Pf.
Neue (F). Formenlehre der lateinischen Sprache.
Band II. Adjectiva, Numeralia, Pronomina, Ad-
verbia, Pripositionen, Conjunctionen, Interjec-
tionen. 3 giinzlich neu bearbeitete <Auflage
von ©. Wagener. 7 und 8 Lieferung. 8vo. 385,
512 Berlin, Calvary and Co.
Ovid. Briefe der Heroiden. Deutsch im Versmasse
der Urschrift von A. Koch. 8vo. 107 pp. Bamberg.
civ, 283 pp. Berlin, Weidmann.
Die Sonnenhelden der Mythologie.
4to. 44 pp. Basel. 1 Mk.
8vo.
Edidit ἃ. Helm-
Leipzig, Teubner.
Buchner. 1 Mk. 50.
Rauch. Gerundium und Gerundivum bei Curtius.
4to. 21 pp. Meiningen. 1 Mk.
Reisig (C. K.) Vorlesungen iiber lateinische Sprach-
wissenschaft. Mit den Anmerkungen von F. Haase.
Theil II. Neu bearbeitet von F. Heerdegen.
Lieferungen 2 ἃ. 8. vii, 97, 155 pp. und Register
von G. Landgraf. lvi pp. 12mo. Berlin, Calvary
and Co. Each 2 Mk.
(Calvary’s philologische
Bibliothek, Band 87 und 88).
Rosshach (A.) und R. Westphal.
musischen Kiinste der Hellenen.
der Rossbach-Westphalschen Metrik. Band 11|.
Abtheilung 2. 8vo. Leipzig, Teubner. 14 Mk.
(Contents: Griechische Metrik mit besonderer
Riicksicht auf die Strophengattungen und die
iibrigen melischen Metra. 3te Auflage von
A. Rossbach. Ixxii, 870 pp.)
Sammlung der griechischen Dialekt-Inschriften von
J. Baunack, F. Bechtel, Bezzenberger, Blass,
Collitz, Deecke, Fick, Meister, Miillensiefen,
Prellwitz. Herausgegeben von Collitz und Bechtel.
Zand III. Heft 4. Hialfte I. 8vo. Gdttingen,
Vandenhoeck und Ruprecht. 2 Mk.
{Contents : Die Inschriften von Aigina, Phole-
gandros, Anaphe, Astypalaia, Telos, Nisyros]
Knidos, bearbeitet von F. Bechtel. 195, 300 pp.,
Schmid (W.) Der Atticismus, in seinen Haupt-
vertretern von Dionysius von Halikarnass bis
und archiologische
Theorie der
Als 3 Auflage
auf den II. Philostratus dargestellt. Band II.
VI. Abschnitt: Aristides. 12mo, 316 pp. Stutt-
gart, Kohlhammer. 5 Mk.
Schreiber (T.) Die hellenistischen Reliefbilder.
Lieferung 2. Folio. 10 plates. Leipzig, Engelmann.
Each 20 Mk.
Statius. lied von Theben, deutsch von A. Imhof.
Mit gelegentlichen sachlichen und kritischen Er-
lauterungen. Theil 11. VII.—XII Buch. 8vo.
153, 328 pp. Llmenau, Schroeter. 3 Mk.
Strack (H. 1.) Vollstaindiges Worterbuch zu
Xenophon'’s Anabasis, 5* vielfach verbesserte
Auflage. 8vo. iv, 155 pp. Leipzig, Hahn. 1 Mk. 20.
Studniczka (F.) Kyrene, eine altgriechische Gottin.
Archaeologische und mythologische Untersuch-
ungen. 8vo. xi, 224 pp. Cuts. Leipzig, Brock-
haus. 9 Mk.
Talbot. Mythologie greeque et Mythologie latine
d’aprés les Travaux de la Critique moderne, 12mo.
xxil, 523 pp. Paris, Lemerre. 2 fr. 50.
Tegge. ateinische Schulphraseologie. Heft 2. 8vo.
94 pp. Berlin, Weidmann. 1 Mk. 20.
Thicle (G.) Quaestiones de Cornifici et Ciceronis
artibus rhetoricis. Dissertation. 8vo. 112 pp.
Greifswald. 1 Mk. 50.
Thommen (E.) Studien zu Thukydides. Inangural-
Dissertation. S8vo. 71 pp. Basel. 1 Mk. 20.
Thucydides. De bello Peloponnesiaco libri VIII. Ad
optimorum librorum fidem editos explanavit E. F.
Poppo. LEditio 111. quam auxit et emandavit
J. M. Stahl. Vol. 1. Sectio 2. 8vo. 260 pp.
Leipzig, Teubner. 3 Mk.
Vries (ὃ. G.) Exercitationes palaeographicas in
bibliotheca universitatis Lugduno- Batavae in-
staurandas indicit de Vries. [Inest commenta-
tiuncula de codice Ciceronis Cat. Mai. Ashburn-
hamensi nune Parisino]. 8vo. 45 pp. Leiden,
Brill. 2 Mk.
Wackernagel (J). Das Dehnungsgesetz der griech-
isechen Composita. 4to. iv, 66 pp. Basel,
Schneider. 1 Mk. 80.
Wessely (C.) Die Pariser Papyri des Fundes von El-
Faijim. 4to. 162pp. Wien. Tempsky. 8 Mk.
[Extr. Denkschriften der kaiserl. Akad. der W%s-
senschaften].
Witrzens (J.) Betonungssystem der griechischen
Sprache. Auf etymologische Grundlage entwickelt.
8vo. vii, 151 pp. Teschen. 3 Mk. 60.
Wohlrab (M.) Die altklassischen Realien im Gym-
nasium. 8vo. ix, 82 pp. Leipzig, Teubner.
Bound. 1 Mk. 20.
Xenophon’s Anabasis fiir den Schulgebrauch heraus-
gegeben von E. Bachof. Band II. Buch 4-7.
Text und Erliuterungen. 8vo. 110, 150 pp.
Cuts. Paderborn, F. Schoningh. 1 Mk. 80.
Historia graeca. Recensuit Ὁ, Keller.
Editio minor. 12mo. xvii, 295 pp. Leipzig,
Teubner. 90 Pf.
-- Memorabilien. Erklirt von L. Breiten-
bach. 6te Auflage bearbeitet von R. Miicke. 8vo.
268 pp. Berlin, Weidmann. 2 Mk. 25.
Zielinsky. Die Marchenkomédie in Athen.
72 pp. St. Petersburg. 2 Mk.
Zimmermann (A.) Kritische Untersuchungen zu
den Posthomerica des Quintus Smyrnaeus. Er-
lauterung zu einer demnichst erscheinenden Aus-
gabe. 8vo. vii, 218 pp. Leipzig, Teubner
4 Mk.
8vo.
The Classical Review
MARCH 1890.
A FUTURE LIFE AS REPRESENTED BY THE GREEK TRAGEDIANS.
Prruaps there is no portion of a nation’s
religious belief that contains so much history
as that which relates to the destiny of man
after death. The degree of civilization
attained, current ethics, national idiosyn-
crasies of moral or physical constitution,
political status, national pursuits—are all
directly or indirectly reflected in a people’s
views as to what is to succeed this present
life of ours ; so that from the peculiar eschat-
ology of a nation at a given period it would be
possible to reconstruct not a little of the
political, social and moral conditions of the
epoch. It is indeed with the nation as with
the individual in this respect as In many
others. If we could know of a man just his
thoughts about the dim hereafter, his hopes
and fears, their influence upon his life and
conduct, or perhaps the absence of influence,
and nothing else—there would be little more
that we should need to know to appraise his
religion, his philosophy, his moral character,
the man himself. ‘ Qualis artifex pereo,’ said
the dying Nero. ‘Deus fio, ut puto,’ sighed
another Roman emperor, in grim satire of
the weary helplessness of his last illness—
such a god! The words are eloquent with
half the story of two lives.
So with the nation: its religion pieces
out its history, its history pieces out its
religion, Now, the Greek of the age of
Pericles has left us in his history, his phil-
osophy, his literature, his art, ample means
of forming some conception of his remark-
able personality. We know how he looked,
what were his salient characteristics, what
were his pursuits and, to a large extent,
what were his thoughts. What then
was the attitude which his_ eager,
NO. XXXI. VOL. IV.
healthful, individualized nature maintained
towards the inevitable hour when he must
leave this world of whose good things he
enjoyed so large a portion, and go out into
the darkness ?
Greek literature presents us with two very
diverse but equally striking pictures of the
after-world. The Homeric poems, and in par-
ticular the eleventh Odyssey, contain the
earlier, the Platonic dialogues, and in par-
ticular the tenth book of the Republic, a
later one. But though no other Greek
writer, except perhaps Pindar, deals with
the question in so direct and interesting a
manner, we should be wrong in taking either
of these views as representing at all nearly
that of the average Athenian of the best age.
Both are probably contributory of a portion,
but of a portion only of the truth ; for
although the thought, even of the average
Athenian of that time, had taken a _philoso-
phic cast that differentiated it considerably
from that of the Homeric Greek, yet we
know that the full Platonic conception was
in definiteness and coherence far in advance of
any that had been previously entertained, and
that even with Socrates himself a far simpler
though not less earnest und spiritual view
was less a matter of belief than of conjecture.
Somewheretherefore between the twoextremes
of primitive superstition on the one hand and
the bold speculation of the philosopher on the
other, should be sought an approximation to
current religious belief. And to this intent
surely one could not do better than investi-
gate the utterances in which the spirit of
that age most truly found voice—its drama.
Socrates and Plato spoke to the philosophic
few, but Aeschylus, Sophocles and Euri-
G
82 THE CLASSICAL REVIEW.
pides to the intelligent many, which in that
great age meant the masses of their country-
men. The soldier of Marathon and Salamis,
the colleague of Pericles in Samos, the
pupil of Anaxagoras, must have been typi-
cal products of their time ; and distinct as is
the impress of a vivid personality upon the
works of each of these three great men, and
with all allowance on the other hand
for traditional sentiment and the stereotyped
allusions of poetical phraseology, we can
hardly doubt that, when they spoke of the
great problems of life and destiny and con-
duct, it was as the majority of their contem-
poraries thought ; that when they were silent
they emphasized a doubt or an indifference
that their audiences would share. They
were teachers, but Athens was παίδευσις
“EAAddos, and they had learnt their lesson at
her knees. What then have they to tell us
of Hellenic anticipations of a future life?
The immortality of influence belongs to the
Greeks as it has belonged to no other nation.
What were their personal hopes and fears
in death and its sequel? what was to become
of that individuality they so highly prized
and reverenced, that youthful intensity of
being, which is spoken of in the Zimaeus as
their dominant characteristic ?
I have said that by the fifth century B.c.
the Homeric conception must have ceased to
be largely representative, but there is little
reason for doubting that the later belief
was essentially the same as the earlier, only
in a different stage of evolution. If there-
fore we would know what survived of it
in later conceptions, we must understand a
little of its general outlines.
At the first glance it is evident that the
early Greek like the early Hebrew had but
the vaguest notions as to a possible future
after death ; indeed on the whole the Ho-
meric Greek had the more definite concep-
tion, such as it was, of the two. It is not
difficult to make a guess at the reason of
this in the case of the Jew, and in that of
the Greek a sufficient general explanation
may be found in his active physical and
moral organization, his buoyant youthfulness
of spirit, his delight in the mere animal
joy of existing in strength and beauty upon
the earth. How could such a creature help
living in the present? For if even the
more immediate future contained the men-
ace of old age and failing strength, the
loss of all that made life to him worth
living, how much more must he have shrunk
with fear and loathing from the thought of
the consummation of that decay in death.
There is a child in all of us, says Plato
(Phaedo, 77 ἘΠ, whose hobgoblin terrors of
the dark must be charmed away: and in
the earlier Greek the child was uppermost
and none knew the potent charms of divine
philosophy to sing it quietly to sleep. So
he projected into the world he dimly
imaged as the abode of the dead the
gloom, the uncertainty, the floating mists,
the joylessness of his own state of mind
concerning it. Being half in doubt whether
the dead exist at all, he imagined a state of
half-existence ; a sort of shadow-play where
the shadow-man sustains feebly the part of
the old joyous self, whose hunting and
wrestling, feasting and fighting are mocked
for ever by the dreary semblance of life and
action, that flits grey and shifting over his
fields of asphodel.
The shade preserves to a certain extent
its personal identity, that is, it retains the
outward form and features of the living
man, his tasks and pursuits, with memory
enough to know what it has lost, emotion
enough to envy the living. But for the rest,
its thought and feeling are as thin and in-
effectual as its bodily constitution. Even
such a mother as Anticlea, who had died
heart-broken with longing for her lost
Ulysses, does not recognize him in Hades,
till he has suffered her to drink the warm
life-blood of the beasts of sacrifice, which
sends flooding back for a brief moment the
tide of memory and affection into the cold
and passionless ghost. ;
This dubiousness as to the actual physical
condition of the departed further affects the
moral side of the question, by the same ‘ con-
fusion of doubt with the object of doubt.’
Our Greek did not know, he could not besure,
that the deeds done in the flesh influenced
for good or ill the lot of the dead. On the
earth, as often as not, the criminal got off
scot-free, and the righteous suffered as the
victims of successful wickedness, or as the
blind puppets of fate. He himself did not
much care to think about dying, but on the
whole he did not see much reason for sup-
posing that, if he were worse off in other
respects, he would be better off in this. So
in the Homeric Hades there is no distinc-
tion between the lot of the virtuous and
the guilty. A few exceptional criminals
indeed, like Sisyphus and Tantalus, suffer
directly, at the hands of the Zeus their ὕβρις
has outraged, Tartarean horrors of punish-
ment; and a few favoured mortals like
Menelaus and Neoptolemus enjoy the
eternal bliss of the Fortunate Islands—
though as a privilege, not as a reward.
Minos and Rhadamanthus are judges
THE CLASSICAL REVIEW. 83
among the dead, but only because they, like
others, continue a semblance of their former
occupations. But their sentence is mean-
ingless and ineffectual, and does not at all
determine the condition of the shade.
Good and bad alike, are only not annihilated :
their common doom is worse—they know
themselves nonentities, νεκύων ἀμενηνὰ
κάρηνα.
How great is the contrast to all this pre-
sented by the story of Er the son of Arme-
nius, by the attitude of Socrates before his
judges, by the whole tone of the Crito and
Phaedo, it is unnecessary to illustrate in de-
tail We may allow that the optimistic
temper of Socrates and the peculiar cireum-
stances of his death naturally of themselves
evolved a hope out of the thought of death ;
that Platonic Ontology and Teleology, re-
quiring the eternal existence of all that is at
all, and denying existence to all but mind,
required as their logical deduction the im-
mortality of soul as soul :—still, granting all
this, it is certain that current theology in
the age of Pericles bore its own traces, even
if scattered and obscure, of a changing atti-
tude towards the thought of death and the
hereafter, just as that there had come upon
the spirit of the age a change in tone and
feeling, which led the way inevitably to
bolder speculation, and was indeed itself the
necessary result of the nation’s growth from
the childhood of the Homeric to the enquir-
ing and fearless adolescence of the Periclean
age.
I.
AESCHYLUS.
* It is often said that the religion of
Aeschylus is a religion of Destiny, a system
where, above and behind even the mightiest
of the Olympian hierarchy, Fate, μοῖρα,
compels the universe to submission, and
makes men the plaything of wanton omni-
potence. And this is true, but with reserva-
tions. Destiny does indeed control man’s
actions, but as the embodiment not of law-
lessness but of law ; man is indeed powerless
to contend with her, but her operations do
not involve the negation of free-will. If
blind infatuation hurries man to destruction,
it is because he failed in that humility and
moderation which he was at first free to
exercise, and because the divine will makes
man the instrument of his own fall by per-
mitting the surrender of all self-control, the
accomplishment of ail dark and impious
designs. And if the incidence of the curse
is sometimes apparently on the innocent
head, that is because the solidarity of the
family—in which Aeschylus believed as
firmly as any modern student of Heredity—
means the inheritance of responsibility ; the
fathers cannot eat sour grapes but the
children’s teeth are set on edge. δρᾶσαντι
παθεῖν. We may not lay on Destiny the
blame of our crimes.
ε a , oS 7 ΄,
ἢ μοιρᾶ TOVTWV, ὦ TEKVOV, παραιτια,
says Clytemnestra in her last hour, fain to
excuse her wickedness. Be it so, replies the
avenging Orestes, but that same Destiny
has wrought thy doom as well.
N a? ΄ ,
και τόνδε τοίνυν μοιρ ἐπόρσυνεν μορον.
Eternal Justice might indeed have used the
adulterous wife as the minister of divine
vengeance on the haughty son of Atreus,
foredoomed by that Thyestean banquet ; yet
not arbitrarily, but through the working of
another law—that one sin leads on to
another, blunting conscience, destroying the
moral sense. Clytemnestra of her own free
will chose the path which led from unfaith-
fulness to murder, so providing a willing
instrument in the hand of the same in-
exorable Fate that in turn demanded her
own destruction. So with Orestes. Obli-
gations both human and divine urge him to
take vengeance for his father. So far his
deed is righteous. Yet inasmuch as he is
the son of Clytemnestra as well as of
Agamemnon, it is sinful, and he must suffer
for the matricide, that the severe integrity
of the moral code may not be impaired, even
in working out its own decrees.
But although Orestes suffers in accord-
ance with the same stern law, ‘ δράσαντι
παθεῖν, that bade him slay his mother, his
punishment in view of that moral obligation
is made instrumental of divine mercy.
“Παθήματα μαθήματα, is the second great
article of the Aeschylean faith, and like the
rebellious Titan, Orestes emerges from his
sufferings strengthened and purified: the
Erinnyes making inquisition for blood are
become the Eumenides, the kindly influences
of a later life that has seen severity justified
by love.
It has been necessary to touch upon the
religious scheme of Aeschylus—which is in-
deed both obscure and complicated in its
details, though hinging on one strong and
central conviction—because a writer’s views
as to the destiny of the soul and a system
of rewards and punishments after death are
not independent of his general belief regard-
a 2
84 THE CLASSICAL REVIEW.
ing the relation between God and man.
After all, what is most of interest is not the
details of allusions to the other world, in
which so much allowance must be made for
poetical imagery, but such questions as:
What is the connexion between this life
and the next? How far do our actions now
affect our condition hereafter? Does the
doer still suffer beyond the grave? Is he
still permitted by suffering to learn? In
order to answer these questions it will
perhaps be advisable to group the passages
for examination under definite headings,
according as they afford typical though not
exhaustive illustration of different important
points of eschatological belief, viz. as relat-
ing to: (a) retribution and compensation
in the future life, with their influence on
conduct; (Ὁ) the maintenance of human
relationships ; (6) intercourse between the
dead and the living; (d) the general condi-
tion of the departed.
I have placed the questions of rewards
and punishments first, because with Aeschy-
lus, as I have already pointed out, the idea
of divine justice and divine wisdom is of
pre-eminent importance. Upon it hangs
his whole theology, a theology in which, as
M. Courdavaux points out in an interesting
study of the poet, the two opposing currents
of contemporary Greek thought met—one
carrying the mind towards higher and purer
ideas of divinity, and the other sweeping
mercilessly back upon the terrors of the
past.
(a) Rewards and Punishnents.—Believing
as Aeschylus did that the two great doc-
trines I have quoted, with that of the
solidarity of the family, sufficiently justify
God’s dealings on earth with men, he does
not look forward to the future life as afford-
ing readjustment of the inequalities, recon-
ciliation of the contradictions, satisfaction
of the frustrated efforts of the present.
This is a modern tendency often enough
carried to an extreme. He believed that
life itself, if we knew how to read it, con-
tains the solution of its own problems. A
firm trust in divine justice and wisdom does
not under these circumstances require that
the second life should essentially differ from
the present, except inasmuch as it is condi-
tioned by a great physical change.
Aeschylus therefore would not look for-
ward with any sense of hope or comfort to
a new order; but inheriting the strong
Greek ‘sub-consciousness of immortality’ he
applies it to the further widening of the
field for the exercise of relentless justice,
‘the most tangible form in which Omnipotence
(‘call him Zeus or by whatsoever title it
please him to accept’) reveals himself to
mortals. There is for instance a passage in
the Eumenides (322) which, after describing
the relentless inquisition for blood of the
Erinnyes, adds, to correct the possible im-
pression that in death at least their victim
might escape,
θανὼν δ᾽ οὐκ ἄγαν ἐλεύθερος.
So a few lines above Night is said to have
borne them
οὖς ἀλαοῖσι καὶ δεδορκόσιν
ποινάν (312).
And again in lines 910-12 :--
μέγα yap δύναται
, > Ν ΄ὔ as ,
πότνι᾽ Epwvs παρά τ᾽ ἀθανάτοις
τοῖς θ᾽ ὑπὸ γαῖαν.
Similarly in the Supplices, 224--227 ---
3OX εὐὐὐς τι") Ν
ΕΣ οὐδὲ μὴ ᾽ν Αἰιδου θανὼν
, , 5 γι / AN
φύγῃ μάταιος αἰτίαν, πράξας τάδε.
3 = ΄ 5 ΄ ᾽ ε ,
κἀκεῖ δικάζει τἀμπλακήμαθ᾽, ws λόγος,
ιν δ΄ 5 a e ». /
Ζεὺς ἄλλος ἐν καμοῦσιν ὑστάτας δίκας.
And among the fragments is one which
speaks of a man as inevitably finding at
last the requital of his sins, whether towards
god or fellow-man :—
μέγας γὰρ ἽΑιδης ἐστὶν εὔθυνος βροτῶν
ἔνερθε χθονὸς
δελτογράφῳ δὲ πάντ᾽ ἐπωπᾷ φρενί.
Instances might yet be further multiplied
to prove that the distinct advance on the
Homeric doctrine, of a clear idea of re-
tribution in the other world, had been
arrived at. But for traces of the correlative
belief in the reward of virtue, which is as
necessary as the punishment of sin for the
perfect ideal of Justice, we seek in vain.
The omission is the more marked by the
contrast offered by his contemporary Pindar.
In the second Olympian, for instance, a
vivid and alluring picture is drawn of the
joys of the just—déaxpuy νέμονται αἰῶνα,
where ocean breezes sigh round the islands of
the blest ; flowers of gold burn on land and
water, whereof they carry garlands in their
hands and wreath their heads with crowns:
while the wicked
, , /
ἀπροσόρατον OXVEOVTL πόνον.
And again in a fragment of the Threnoi:
‘For them shines the might of the sun
below, when here it is night ; and meadows
of red roses skirt their city, shaded with
THE CLASSICAL REVIEW. 85
incense-trees and orchards laden with golden
fruit. And some delight themselves with
horses and wrestling, and some with draught-
playing, and some with lyres, and around
them fair-flowering all plenty blooms. And
a delightsome smell is spread about the
place, where they mingle ever all goodly
spices in the beacon-flame upon the altars
of the gods.’
There is nothing like this promised in the
pages of Aeschylus, and yet tradition says
that he like Pindar was a Pythagorean (Cic.
Tusc. II. 10). ‘Non poeta solum, sed
etiam Pythagoreus: sic enim accepimus.’ If
therefore, as we are told, Pindar’s peculiar
views about the world to come are of Pytha-
gorean origin, it is the more surprising that
Aeschylus has nothing to say of the happi-
ness in store for the pure and holy of life.
But perhaps the poet, who believed (4g. 797)
that Justice may often be found in the
smoke-dimmed cabins of the poor, when the
ὕβρις of the rich has banished her from their
gilded palaces, revolted from the anticipa-
tions which promised the futurity of bliss in
those islands of the west only to the great
and strong. Without wealth and prosperity
the virtues of the μεγαλόψυχος hero of”
Pindar could have had no room for develop-
ment, and indeed the value of wealth in
this connexion is certainly hinted at in the
second Olympian, though the exact inter-
pretation of the passage is obscure. After
all, the difference between the two is mainly
in the point of view. With both the future
continues the essential characteristics of the
present, but circumstances and _ natural
temperament lead the two poets to bring a
different side of the belief into prominence.
With Pindar to be rich and prosperous was
in all probability to be virtuous, provided
always the due sense of moderation and
submission to the gods kept men secure from
divine envy. ὕβρις is with him the dreadful
exception ; so he delights to paint the rewards
of the future. But with Aeschylus ὕβρις is
all too common ; the rich too often fall from
overweening confidence in their prosperity.
Men must be warned by the continual
reminder that Justice though halt of foot is
unfailing in her pursuit and, though she
may not overtake her victim in this life,
that there is the life beyond where escape is
hopeless.
If Aeschylus could have freed himself
from the dreary influence of the past and
of his own almost exaggerated moral sense,
his real belief in divine justice and wisdom
would no doubt have found room to expand
into some comforting hope for the sorrowful,
the suffering, and the oppressed, perhaps
too into a belief that in the future life as
well, human character might be sublimated
by suffering. But if his belief ever worked
out its logical conclusion, he did not give it
to his fellows, certainly not in his surviving
works. Practically with him the other
world exists, at least as a felt influence on
conduct, only for the guilty.
‘Victurosque dei celant, ut vivere durent,
Felix esse mori—’
said Lucan, stung with that fierce depres-
sion that seems the bane of a complex
civilization. Perhaps the thought was in
the mind of Aeschylus that for a yet loftier
purpose the gods conceal the happiness of
death, that chastening fear and patient
acceptance of discipline might merit its
rewards, and prepare fit recipients of the
gift of immortality.
(ὁ) Permanence ef Human Interests and
Relationships.—That the dead retain a dis-
tinct personality is part of the rudest forms
of belief in an after-life, and in Greece indeed
belongs rather to the simple popular beliefs
than to the philosophical systems, which tend
either in the direction of Metempsychosis
or of reabsorption of the individual in the
Universal Soul. Neither of these modes of
extended existence can be described as per-
sonal immortality, though they may satisfy
a belief in the immortality of the life-
principle. In Aeschylus there is no trace
of either of these hypotheses. Pythagorean
though we are bidden to consider him, not
only personal characteristics remain, but
the more transient circumstances of rank
and influence. Darius is still in Hades
δυναστᾶν δυνάστης, reigning as king among
the dead. And if Agamemnon was without
such honour, it was only because he had died
in no kingly fashion, shamed and dis-
honoured, without due rites of worthy
burial. Would that he had died on the
scene of his triumphs, though in a foreign
land, yet in a soldier’s grave, sing the chorus
of the Choephoroe—
φίλος φίλοισι τοῖς
ἐκεῖ καλῶς θανοῦσιν
κατὰ χθονὸς ἐμπρέπων
σεμνότιμος ἀνάκτωρ
πρόπολός τε τῶν μεγίστων
χθονίων ἐκεῖ τυράννων,
βασιλεὺς γάρ ἦν, ὀφρ᾽ ἔζη,
μόριμον λάχος πιμπλάντων
χεροῖν πεισίβροτόν τε βάκτρον.
Choe. 346-354.
86 THE CLASSICAL REVIEW.
But not these outward trappings only sur-
vive the transition. Moral characteristics
still cling to the bodiless ghost—fierce un-
dying revenge, as well as the kindly feelings
sprung from tender human relationships :—
τέκνον, φρόνημα τοῦ
θανόντος οὐ δαμάζει
πυρὸς μαλερὰ γνάθος.
Choe. 510.
The wrath of Agamemnon (Choe. 285) can
make itself felt among the living in disease
and madness, the assaults of the Furies and
dire dreams, as long as he is unavenged ; he
can distinguish between the offerings of
filial affection, however humble, and the
perfunctory sacrifice of unholy hands. The
lament of Electra (Choe. 364) is a more pre-
cious offering than gold, and its soothing
influence has become a helper of his children
with the wrathful dead :—
τῶν δὲ κρατούντων
χέρες οὐχ ὅσιαι στυγερῶν τούτων"
παισὶ δὲ μᾶλλον γεγένηται.
Choe. 370-371.
The Humenides and Persae afford many
similar instances. But here again we some-
what wonder that it is still the gloomy side
of the case that is prominent—nothing of
reunion with those we have loved and lost
to soften separation, no thought that we go
to them though they return not to us. A
strain of fresh human sympathy there surely
was, tender and pitiful, in the poet who has
left us the pleading eyes of Iphigenia and
her dumb lips that should have moved pity
with memories of sweet girlish songs. It
was there, but it makes no sign to soften
the terrors of futurity.
(c) Communication between the Living and
the Dead.—In the seven extant plays of
Aeschylus there are two apparitions of the
dead, and numerous allusions to ghostly
visitations as well as to the phantom forms
of dreams. Both the eidolon of Clytemnestra
and of Darius appear visibly on the stage
and address themselves to the Chorus. They
are therefore actual shades exercising their
privilege of re-visiting the earth and holding
intercourse with the living. The ancients
seem to have believed in a very close con-
nexion between these waking visions and
the forms we see in dreams, which they
considered to be equally faithful revelations
of the spirit-world, though different in kind.
Some anthropologists go so far as to say
that a belief in an existence after death
actually arose in primitive peoples purely
from an attempt to explain dreams of the
departed. However that may be, the Greeks
had a firm belief in the power of the dead to
appear to the waking senses in a vision
(ὕπαρ) ; and when the bodily senses are laid
to sleep, they believed that the mind by its
own peculiar faculties, whose activity is in-
creased in proportion as they act inde-
pendently of the body, perceives the spiritual
inhabitants of the invisible world with which
it has itself more intimate connection than
with the visible.
ὁρᾶτε πληγὰς τάσδε καρδίας ὅθεν;
εὕδουσα γὰρ φρὴν ὄμμασιν λαμπρύνεται---
cries Clytemnestra to the sleeping Furies ;
and though she describes herself as ὄναρ, yet
it is evident that there is present before us
all that is left of the dead queen who, ever
hungry for revenge, and scorned and flouted
in Hades, rises to the upper air to goad on
the sleeping ministers of her wrath.
This idea of the duality of human nature,
body and spirit, is traceable in Pindar, and
reaches a definite conception in EKuripides.
For instance in the Threnoi occurs the
following interesting fragment—
3 7
“Ολβίᾳ δ᾽ ἅπαντος αἴσᾳ λυσίπονον τελευτάν
καὶ σῶμα μὲν πάντων
σθενεῖ,
Ν 3 δι ’ A: Μ' u Ν ΄
ζωὸν δ᾽ ἔτι λείπεται αἰῶνος εἴδωλον: τὸ γάρ
ἐστι μόνον
> ΄“- ΄ Ἂς , / 3s
ἐκ θεῶν: εὕδει δὲ πρασσόντων μελέων, ἀτὰρ
εὑδόντεσσιν ἐν πολλοῖς ὀνείροις
δείκνυσι τερπνῶν ἐφέρποισαν
κρίσιν---
΄ /,
ἕπεται θανάτῳ περι-
χαλεπῶν τε
containing a clear assertion of the antithesis
of body and spirit, of the alternatives of
happiness and woe in the life to come, and
of the prophetic perception when ‘we are
laid asleep in body and become a living soul.’
This last belief is also evidently alluded to
in the next line to the passage I have already
quoted from the Humenides,
ἐν ἡμέρᾳ δὲ μοῖρ᾽ ἀπρόσκοπος βροτῶν,
evidently implying that at night destiny
unveils herself to men. In Euripides the
belief in this duality of nature and in the
superiority of the spiritual to the material
essence leads towards the conclusion that
the greater and better portion of human
existence is what we now call death, since
then the spiritual escapes the trammels of
the body ; the gain of which may be argued
from this very fact, that even the partial
loosening of the fetters in sleep permits the
soul to penetrate the unseen. Thus the
natural awe that springs from the mys-
teriousness of dreams tends to give them
THE CLASSICAL REVIEW. 87
value, and to carry the mind with them
back to the unseen from which they seem
to spring. And that in turn gives them
fresh importance as glimmerings and _ in-
timations out of the spiritual world to which
our own better selves, ‘such stuff as dreams
are made of,’ belong, and which ‘ rounds our
little life with sleep.’
To such a feeling we must ascribe the
frequency of allusion to dreams in the plays
of Aeschylus, and the importance he attaches
to such visitations,
“ lal ’ὕ
πτεροῖς ὁπαδοῖς ὕπνου κελεύθοις.
(d) The general condition of the Dead.—
There is little to add to our account of the
other world as Aeschylus shows it to us.
It hasits own ministers of justice who carry
out the principles of earthly equity ; its own
stern deities, once hostile now subservient
to the all-accomplisher Zeus Teleios, who
in the other world works out the decrees
that are promulgated in this: rewards and
compensation there may be, retribution there
assuredly is: the dead retain their earthly
passions and attributes : they have power to
exact from their kindred the satisfaction of
their wrongs by troubling their peace, and
they can make their influence felt upon the
pious and the dutiful, as kindly δαίμονες,
shielding from harm, and bringing peace and
prosperity upon those who have deserved
their goodwill. But they in turn are depen-
dent upon the living for such poor comfort
and peace as the nether world can give.
Due performance of their burial rites and
worthy veneration of their memory are essen-
tial to securing repose and veneration among
the shades. This is all that is revealed of
comfort—release from bodily suffering for a
Philoctetes—
3 , a
Θάνατε Παιάν, μή μ᾽ ἀτιμάσῃς μολεῖν"
μόνος γὰρ εἶ σὺ τῶν ἀνηκέστων κακῶν
> > ty, n
ἰατρός, ἄλγος δ᾽ οὐδὲν ἅπτεται νεκροῦ.
Frag. of Phil. apud Stobaewm—
and for all rest, deep and endless and in-
different,
παροίχεται πόνος,
παροίχεται δὲ τοῖσι μὲν τεθνηκόσιν
τὸ μήποτ᾽ αὖθις μηδ᾽ ἀναστῆναι μέλειν----
Ag. 501, cf. Choe. 510---
or at best a faint reflection of the past.
It is a gloomy conception. Aeschylus him-
self, as Dr. Westcott points out, half inclines
to think that a secondary positivist im-
mortality in our children is more consolatory
(Choe. 512). Like Homer he realized the
nobility of our present powers, and felt in
that intense Greek life that the fulness of
existence is exhausted here. Here and now
God sees and judges ; itis only the execution
of His sentence that is sometimes delayed.
It is enough for us to walk circumspectly
without prying too curiously into the coun-
sels of omnipotence. Deeply and truly re-
ligious though Aeschylus was, his religion
was one of terrorism, and could not fail to
turn away thought from the future with a
sentiment of vague alarm. He is essentially
the prophet of the Greek drama, a religious
reformer whom the age brought forth to
meet the widely felt demand for something
that should harmonize the old stories of the
mythologers with the abstract and elevated
ideas that began to be formed of deity.
Orphism and the mysteries are products of
the same movement, and indeed it is more
than probable that Aeschylus himself fell
under the spell of this influence. The old
tradition that he was accused of profaning
the mysteries by allusions in his plays is not
so unlikely as it would at first appear. That
he was ever actually initiated we do not
indeed know, but his connexion with Eleusis
makes it at least probable. And if it be
objected that so religious-minded a man was
not the one to be guilty of an act of pro-
fanation, on the other hand it may be argued
that the very intensity of his zeal was likely
to have led him to wish to share with his
fellowmen such salutary teaching as he might
with his quickened discernment have gathered
from the Eleusinian rites. Like most great
reformers he was a little one-sided. The
truth which it was his special mission to
preach, he was prepared to enforce some-
what at the expense of others almost equally
important. And he was right. If the
brutalizing influences of the old religion
were to be got rid of, when man worshipped
a reflected self, with all his own weaknesses
and passions glorified as divine, the first
truth to be enforced with inexorable severity
was that of the stern and absolute majesty
of a god of purer eyes than to behold iniquity.
Men must realize that fact. Later philosophy
might then teach ‘the All-wiseis the All-loving
too,’ and certain of their own poets remind
them that they also were His offspring. With
such a thought men might at last venture
to look with hope as well as dread into the
future life, and seek to form a wider and
more satisfying conception of its character.
But the stern teaching of Aeschylus was
first needful, that they might understand
that love need not be weakness, and that
severity to sin may prove the highest form
of mercy.
88 THE CLASSICAL REVIEW.
ΠῚ:
ΞΟΡΗΟΟΘΠΕΒ.
And if Aeschylus is pre-eminently the
prophet, the teacher of the Greek drama,
Sophocles is first and foremost its supreme
artist, the tender and refined, appealing to
the feelings rather than to the intellect, toa
sense of what is moving and pathetic rather
than of what is sublime and terrible. In
literature they stand to each other in some-
what the same relation as Pheidias and
Praxiteles in sculpture. Praxiteles, says
Diodorus, portrayed in his marble τὰ τῆς
ψυχῆς πάθη, that is, emotions, moods, feelings,
sensations. But Pheidias deals, not with the
πάθος, but with the ἦθος of man—his general
abiding characteristics, what he is, not what
he feels (cf. J. E. Harrison’s Studies in Greek
Art). Aeschylus then is the Pheidias of
Greek tragedy, Sophocles its Praxiteles, and
their work respectively ἠθικός and παθητικός,
ethical and pathetic in a limited English
sense, and something more. And as we have
seen in Aeschylus, this large ethic treatment
of human nature in its subjection to law
and its connection with the divine, evincing
itself in a stern and uncompromising con-
ception of the future state, so in Sophocles
it seems @ priort probable that a more
human sympathy with the pathos of our
earthly life would extend over his forecast
of ourultimate destiny its softening influence,
and induce him to dwell on the hopeful
rather than the melancholy aspects of our
passage to another world.
Asa matter of fact Sophocles does not
dwell with insistence upon the question. ‘The
grim traditions of the past, the gloomy
Chthonian worship in which Aeschylus seems
to have taken a certain stern delight, had no
attractions for his more polished artist
nature. With him the thought that death
has its compensations, even if mainly negative
ones, is frequent, while the pains and penal-
ties of the other world are seldom alluded to.
He is concerned not to vindicate principles
so much as to write a perfect play, which
shall produce in his hearers that serious
pleasure and pleasant pain which the
moving and the pathetic treated with the
highest art evoke. Not that there is no
moral purpose in the plays of Sophocles : far
from it. But on the whole he treats his
matter objectively, and while Aeschylus
speaks ὧν propria persona through his
characters, Sophocles with less inspiration
but better art makes his characters speak
their own thoughts.
The plays that on the whole are most
fruitful of the information of which we are
in search, are the Antigone, the Ajax, and
the Oecedipus Coloneus, in all of which the
chief characters stand face to face with
death—in the first two cases as the object of
deliberate choice, in the other as the des-
tined goal of long and weary vicissitudes.
How then do they regard it? ‘No one,’
says the Chorus in the Antigone ‘is so foolish
as to be enamoured of death,’ 220. An-
tigone, as Prof. Jebb points out, makes a
perfectly sane choice in meeting death, only
because the alternative was to neglect a
sacred duty. The utterances of this play
ought therefore to be more weighty and un-
biassed than those of the shame-stricken
Ajax, eager to hide his dishonoured. head
in night, or of the travel-worn Oedipus,
longing for easeful death to rid him of the
burden of a life that had been from birth
mocked and flouted by Fate.
Importance of Burial.—The first question
that the Antigone raises is that of the im-
portance to the dead of burial. When the
decree of Creon is first mentioned, Antigone
says that Eteocles, whom the king has
suffered to be duly buried, is τοῖς ἔνερθεν ἔντι-
pov νεκροῖς, a state of things which is clearly
traceable, not to superior merit, but to his
having secured the reverence of the dead by
the reverence of the living. For there is a
tendency among the nether folk, like that
prevalent in the upper world, to trample on
those who are down, so that those who
neglect or refuse to perform the burial rites
of the dead are not only found guilty of
failing in the most sacred duties of kinship
and nationality, but of the heartless cruelty
of robbing the dead of all hope of honour
and peace in Hades.
The seriousness of such a crime is illus-
trated by the severe sentences that were
passed upon Athenian generals who failed to
gather up the slain for burial, as for instance
after Arginusae, and by the conduct of Nicias,
who on one occasion turned back with victory
within his grasp, on finding that the bodies
of two Athenians had been left behind un-
buried, preferring to forego the victory to
wronging his dead countrymen.
Caelum tegit qui non habet urnam,
said the Roman poet, and Socrates reminded
his sorrowful friends that it was not himself
they would bury, who would be far away out
of their reach. But in the popular Greek
belief burial was an important event in each
man’s existence, one on which his future
fate largely depends. The devotion of
Antigone was therefore animated by the
THE CLASSICAL REVIEW. 89 -
highest of motives, her brother’s happiness
not in time but in eternity. Seen in this
light, the wrangling over the body of Ajax,
which to modern taste seems a fault in the
construction of an otherwise fine play, is no
longer an inti-climax but an integral part
of the hero’s story, full of point and sig-
nificance.
It is evident therefore that in the view of
Sophocles the dead are in an important
respect dependent on the living for their
future state. We go on to ask how far
this is further influenced for good or evil by
their actions in the earthly life.
Rewards and Punishments. — ‘ Justice,’
says Antigone, ‘has her dwelling with the
gods below, and ruthlessly exacts her due
from men.’ Creon, who has retained in this
world the dead Eteocles after he has become
the proper subject of the nether kingdom,
must surrender his own son Haemon—
[Δ cal > Ν 5 4
VEKUV VEKP@wV ἀμοιβὸν ἀντιδούς.
But does Justice act against the dead them-
selves? This is tacitly implied perhaps, but
nowhere in Sophocles is it insisted upon as
it is in Aeschylus. Indeed he introduces a
most disturbing element into his conception
in the justificatory influence of initiation
into the Mysteries. In a fragment of an
unknown play, he says
ὡς τρισόλβιοι
κεῖνοι βροτῶν ot ταῦτα δέρχθεντες τέλη
μόλωσ᾽ ἐς “Avdov: τοῖσδε γὰρ μόνοις ἐκεῖ
ζὴν ἐστί, τοῖς δ᾽ ἄλλοισι πάντ᾽ ἐκεῖ κακά.
There is a similar statement in the Threnoi
of Pindar.
The Mysteries—It is now generally
accepted as proved that the Mysteries,
although originally mainly symbolical of the
operations of nature, with her alternations
of life and death, winter and summer,
represented under the beautiful myth of
Demeter and Kore, were further applied to
the story of our common human destiny.
The evidence of this from what we know of
the rites and ceremonies of the Mysteries,
from the study of comparative mythology,
from vase-paintings and scattered allusions
cannot now be entered into. But it is
sufficiently evident that the analogy between
the history of the corn of wheat buried out
of sight in the dark earth only to spring up
into an ampler and richer life, with which we
are so familiar under St. Paul’s treatment,
had already been seized by the Greek
Mystics, perhaps appropriated through the
Egyptians out of the symbolism of an ageless
past.
How far however the teaching connected
with this analogy conveyed hope and conso-
lation to the adepts we cannot say, except
that the merits of the purifications and
sacrifices of the ceremonies seem to be car-
ried into the great mystery of all as justi-
fying a right to joys and privileges beyond
the reach of the profane multitude. But
that the Mystae had any formulated doctrine
of a future state is altogether doubtful.
Indeed it is on the other hand almost
certain that the aim of the initiation was less
to impart definite knowledge, than to pro-
duce a state of mind susceptible of certain
impressions, whose character was directed by
the scenic surroundings and dramatic charac-
ters of the rites. Dr. Mozley in an inter-
esting lecture on ‘Pagan Conceptions of a
Future Life’ points out that the very fact that
such a belief was taught suggests a want of
real conviction. The possessor of true and
consolatory belief would never submit to keep
it shut up in his own soul, for true belief
rests on our common human nature, human
conscience and human reason. ‘The crowd’
he says, ‘ played with the imagery of another
world, but it had no true place as a truth in
their hearts; nobody lived for it; the old
Pagan imagery was an enormous advance
upon belief.’
In so far then as the Mysteries concerned
themselves with the future life, they afford
a further illustration of the fact that the
Greeks as a nation believed that, whatever
in the main might be our life here, it was
continued or reflected beyond the grave: that
the Mystae of Eleusis carried with them
their exclusive holiness, and might hug
themselves with the anticipation of the same
vague and intangible benefits in that other
world that accrued to them in this. Such
a belief Sophocles shared.
That he did not however consider it
incompatible with an earnest conviction of
the infallibility of divine justice, that he
could hold such a doctrine of predestination
at the same time that he asserted that
Justice is essentially an infernal deity,
proves how dim and floating all these beliefs
were. There is a similar inconsistency in
what he says about the fate of Polyneices.
The dying Oedipus, with a curse whose
binding nature would have been generally
admitted, consigns his rebellious son in death
to
τὸ Ταρτάρου
στυγνὸν πατρῷον ἔρεβο-----
and yet the sisterly affection of Antigone has
power to secure him peace and privilege in
90 THE CLASSICAL REVIEW.
Hades. That justice would not intervene to
separate the lot of the two brothers in
accordance with their different merits, is
very clear. For Creon suggests to Antigone
that Eteocles might reasonably object to
receiving only the same honour as the
wicked. To which Antigone replies that no
one knows whether feuds be not made up in
the world below, so that Polyneices might
then think it fit and pious that Eteocles
should be honoured.
οὔτοι ποθ᾽ οὗχθρός, οὐδ᾽ ὅταν θάνῃ, φίλος,
replies Creon, and he is allowed the last word.
There would be no room or occasion for such
feelings if Eteocles were in bliss and Poly-
neices in a state of punishment.
Consolatory Allusions. —We gather there-
fore that Sophocles is in teaching less ad-
vanced than his elder rival. But while the
terrors of the other world are far less
prominent with him than with Aeschylus,
there are on the other hand not wanting
hints that there are some aspects of death
which are half consolatory to the suffering
and distressed.
The whole episode of the death of Oedipus
at Colonus is treated with tender reverence,
as the passing of an old blind king into peace
from a life that had grown wearisome: and
in clear and unequivocal terms Theseus
forbids his children to lament—
παύετε θρῆνον, πᾶιδες: ἐν οἷς γὰρ
χάρις ἡ χθονία ξυναπόκειται
πενθεῖν οὐ χρή" νέμεσις γάρ.
0.C. 1750.
Several passages speak of death as sleep :
καλῶ δ᾽ ἅμα
πομπαῖον Ἑρμῆν χθόνιον, εὖ με κοιμίσαι, Says
Ajax before his suicide, and the same image
occurs in 0.7. 961 and 0.C. 1577. So An-
tigone uses the epithet παγκοίτας of Hades.
Or again death is the saviour from all
troubles—érikoupos τῆς λύπης, Ο.Ο΄. 1220, as
in the prophecy to Heracles—
μόχθων τῶν ἐφεστώτων ἐμοὶ
λύσιν τελεῖσθαι.
τὸ δ᾽ ἣν dp’ οὐδὲν ἄλλο πλὴν θανεῖν ἐμέ---
Trach. 1171.
And for Philoctetes and Ajax death is the
Healer, long desired, much entreated :—
ὦ Θάνατε, Θάνατε, πῶς ἀεὶ καλούμενος
οὕτω κατ᾽ Hap, οὐ δύνᾳ μολεῖν ποτε ;
Phil.
The Survival of Human feelings and of Hu-
man Relationships.—The passages quoted
about the undying enmity of Polyneices and
Eteocles have already given proof that
human passions and relationships survive
death. So too Oedipus, in his bitter shame
and remorse, asks how he will be able to
meet in Hades the father and mother he
so foully wronged—
ΘΈΕΙΝ, Ν 3 ΞΟ) oF / /
eyw yap OUK οἷὸ ομμασιν ποιοις βλέπων
i? >on AQ > ¢ ,
πάτερα TOT Gv προσειὸον εις Αιδου μολών
οὐδ᾽ αὖ τάλαιναν μητέρα. . . ΟΣ ΠΕ ΠΕΡ
Memory and remorse will survive therefore
—perhaps sufficient punishment.
But there are kindlier aspects of the same
fact. Ismene (Ant. 65) trusts that the dead
will see extenuating circumstances in her
position, and Antigone is assured that the
gratitude of her kin, which after her own
death will irradiate with love the dark house
of death, will be worth all her suffering-—
καλόν μοι τοῦτο ποιόυσῃ θανεῖν.
{A ? > aA / 7 [2
φίλη μετ᾽ αὐτοῦ κείσομαι, φίλου μέτα,
ὅσια πανουργήσασ᾽: ἐπεὶ πλείων χρόνος
Δ “A a lal
ὃν δεῖ μ᾽ ἀρέσκειν τοῖς κάτω τῶν ἐνθάδε.
ἐκεῖ γὰρ ἀεὶ κείσομαι: σοὶ δ᾽ εἰ δοκεῖ,
a “ ᾽
τὰ τῶν θεῶν ἔντιμ᾽ ἀτιμάσασ᾽ ἔχε.
=
Ant. T2—76.
Most beautiful and touching of all however
is the address of Antigone to the rock-hewn
chamber which is to be her living tomb.
Fully aware of all she loses, and yearning
surely, though in silence, for Haemon’s love,
the significant μέντοι carries the assurance οὗ
her hope :—
ὦ τύμβος, ὦ νυμφεῖον, ὦ κατασκαφὴς
οἴκησις ἀείφρουρος ot πορεύομαι:
πρὸς τοὺς ἐμαυτῆς ὧν ἀριθμὸν ἐν νεκροῖς
a Q
πλεῖστον δέδεκται ἸΤερσέφασι ὀλωλότων"
ὧν Now bia’ γὼ καὶ κάκιστα δὴ μακρῷ
κάτειμι, πρίν μοι μοῖραν ἐξήκειν βίου.
ΕῚ “ / / 3. 59 > ΄ /
ἐλθοῦσα μέντοι κάρτ᾽ ἐν ἐλπίσιν τρέφω
, Ν WA , δ ‘ /
φίλη μὲν ἥξειν πατρί, προσφιλὴς δὲ σοί,
μῆτερ, φίλη δὲ σοί, κασίγνητον κάρ α---
3 Ν ie > / ε a DN
ἐπεὶ θανόντας αὐτόχειρ ὑμᾶς ἔγω
+ Je 3
ἔλουσα κἀκόσμησα κἀπιτυμβίους
χοὰς ἔδωκα.
Ant. 891.
And besides this compensation Prof. Jebb
points out that Antigone dies with the
belief that the question which the poet has
left unanswered as to her personal guilt,
ὅσια πανουργήσασα, Will in the other world
be solved—
ἀλλ᾽ εἰ μὲν οὖν τάδ᾽ ἐστὶν ἐν θεοῖς καλά,
παθόντες ἂν ξυγγνοῖμεν ἡμαρτηκότες--
Ant. 925-6.
The gods have let her die for obedience to
THE CLASSICAL REVIEW. 91
their unwritten laws, and the why, that has
tortured many a soul since hers, goes with
her for its answer into the tomb.
There is this undertone of melancholy
running throughout the plays of Sophocles,
and if he dwells somewhat fervently upon
the rest, the healing, the reunion in the
other world, it is less from a strong appreci-
ation of these joys as such than from a
sense of the struggles, the suffering, the
separations of our life here, which force us
to ery aloud on death for comfort. Such
happiness as it can offer us is after all only
relative. To the prosperous and contented
death has nothing to offer. It is the cold
dreariness of night: but if we are only
longing to be asleep, what matter ?
ἹΠΠῸ
EURIPIDES.
And what of Euripides—the realist, the
philosopher, or rather perhaps the sophist,
the sceptic, the disciple of Anaxagoras, the
friend of Socrates? Far from avoiding the
question of our future destiny, that is just
the kind of speculation that fascinates him.
The fragments of his lost plays alone are
full of allusions to it, and in most of those
passages there is a real earnestness that does
not always characterize his philosophical
disquisitions. Many of them remind us of
Socrates, the δεινὸς σοφιστής, to whom he
seems to allude in the Hippolytus, and
though Euripides does not seem to have
been actually his disciple, there can be little
doubt that as his friend he was conversant
with most of his teaching. For instance:
΄ 3. ἐπί, 2 9A Af? a 7, a
τίς δ᾽ οἶδεν εἰ ζὴν τοῦθ᾽ ὃ κέκληται θανεῖν
Ν lel x / > , ie) c al Lal
τὸ ζῆν δὲ θνήσκειν ἐστί, πλῆν ὁμῶς βροτῶν
a ε 4 ε Ἄν ἐπὶ /
νοσοῦσιν ot βλέποντες, οἱ δ᾽ ὀλωλότες
3 Ν “- 3 Ν ’ 7,
οὐδὲν νοσοῦσιν οὐδὲ κέκτηνται κακά.
Frag. 967.
Socrates’ last words to his judges might
paraphrase the earlier part of the passage,
and the thought that seems to lie behind
his reminder to Crito of the debt to pay the
god of healing is parallel to the latter part.
So too there are echoes of the great teacher’s
saying, Bios μελέτη θανάτου, Which must have
haunted Euripides’ mind, so often does he
revert to the thought of death. In a
fragment of the Alewandros he says :
4, Ν A
πάντων τὸ θανεῖν: τὸ δὲ κοινὸν ἄχθος
μετρίως ἀλγεῖν σοφία μελετᾷ.
Cicero, in his Third Tusculan, chap. xiv.,
translates a lost passage of Euripides, where
Theseus says that, in obedience to the pre-
cepts of some philosophers, he constantly
meditated on death and other possible forms
of evil, so that if any disaster should
suddenly befall him he might not be over-
whelmed by the unexpected blow. On which
Cicero makes the following comment : ‘Quod
autem Theseus a docto se audisse dicit, id de
se ipso loquitur Euripides. Fuerat enim
auditor Anaxagorae, quem ferunt nunciata
morte filio dixisse, “Sciebam me genuisse
mortalem.’’ Quae vox declarat iis esse haec
acerba, quibus non fuerint cogitata.’ That
knowledge, if we could but attain to it, of
our future state would have an appreciable
influence upon our actions, he fully believed ;
half our foolishness and error and ‘ lovesick
fondness’ for our present life being due to
our ignorance of it.
πᾶς δ᾽ ὀδυνηρὸς Bios ἀνθρώπων
κοὐκ ἔστι πόνων ἀνάπαυσις"
ἀλλ᾽ ὅ τι τοῦ ζὴν φίλτερον ἄλλο
σκότος ἀμπίσχων κρύπτει νεφέλαις.
δυσέρωτες δὴ φαινόμεθ᾽ ὄντες
τοῦδ᾽ ὅτι τοῦτο στίλβει κατὰ γῆν
du ἀπειροσύναν ἄλλου βιότου
κοὺκ ἀπόδειξιν τῶν ὑπὸ γαίας"
μύθοις δ᾽ ἄλλως φερόμεσθα.
Hipp. 190—197.
The last words are significant. The old
traditions and explanations are μῦθοι, and
the mystery remains. His position is
agnostic ; but not, as is so often the case,
with the half-conviction that not only can
we know nothing, but that there is nothing
to know. Rather that the best part of all
knowledge lies behind the veil, the privilege
of the spiritual self that will at last penetrate
reality :—
\ gn en, A A 29 ,
τὸ ζὴν γὰρ ἴσμεν: τοῦ θανεῖν δ᾽ ἀπειρίᾳ
πᾶς τις φοβεῖται φῶς λιπεῖν τόδ᾽ ἡλίου.
Traces of Orphism.— Some writers have
expended much learning in attempts to
prove the presence of Orphism in the plays
of Aeschylus and Euripides. The amount
of truth however in the statement is of the
kind continually exaggerated by writers
with a hobby. The teaching of Orphism
was not, like that of the Mysteries, confined
to the initiated, and might therefore with
its pantheism, its doctrine of metempsy-
chosis, its exaltation of a life of ascetic
purity, well have had attractions for minds
like those of Aeschylus and Euripides eager
in search of something that should remedy
the spiritual poverty of the national religion,
93 THE CLASSICAL REVIEW.
and ready to assimilate floating elements of
truth from whatever source they sprang.
But to say that Aeschylus and Euripides
were Orphists would assuredly be beside the
mark. Each appropriated from Orphism
what answered the special craving of his
mind, and let the rest go by. There is
in Aeschylus no trace of pantheism—on the
contrary, he was a polytheist nominally,
practically a monotheist—no trace of a
doctrine οὗ metempsychosis. But the Orphic
spirit is there, the deep reverence for the
unseen powers, the love of purity and right-
eousness for their own sake, the penetrating
enthusiasm, that were the notes of its
presence.
It is something quite different that at-
tracts Euripides. It is the ascetic separa-
tism of mystic cults in which his proud
disdain of the mob finds satisfaction ; and
the purity of body and soul that Artemis
loves in the young Hippolytus typifies
perhaps half-unconsciously to the sceptic
poet the purity of heart that gives insight
into the mysteries he longed to probe.
These things Euripides doubtless found
attractive in the purer forms of Orphism.
But his was a mind that could satisfy itself
with the whole of no fixed belief of his
times. There is in him none of the longing
of the essentially pious Sophocles for
confidence and devotion. It is in Sophocles
and Aeschylus that we find the real Greek
religion, ‘tantum quae potuit suadere ma-
lorum,’ glorified by the first rays of the
rising sun of philosophy. But Euripides
parted with it to share with Socrates and
Plato a longing to give voice to the common
aspirations of humanity, prompted by the
spirit of enquiry so strong upon that age
to throw away the acquiescence and timidity,
if not indifference, of the past, and to seek
an explanation of the individual in the uni-
versal, of man in God, and of the present
life in the life to come.
Influence of Philosophy.—Wonder, says
Aristotle, is at the bottom of all philosophy.
And so when under the guidance of Pro-
tagoras and of all the greater Sophists,
including Socrates himself, the attention of
the thoughtful came to be concentrated upon
man himself, as the only object in this
world of fleeting phenomena which can be
studied with some advantage in the attain-
ment of truth, the feeling of wonder and
admiration at ourselves, so secretly and
wonderfully made, imperceptibly grows into
the timid belief that the great Demiourgos
did not mean his highest work merely for
swift destruction. And the pupil of Anax-
agoras who had talked with Socrates must
have been led to the deduction from his
master’s teaching which his master himself
did not perceive—that, with his hypothesis of
mind governing the world, all things must
be directed εἰς τὰ ἄριστα :—
> , lal > ‘\ > os /
ει θεοί τι δρῶσιν αισχρον OUK εἰσιν θεοί,
he says. If therefore other laws, of which
we have no knowledge, do not intervene,
there is a reasonable probability that some-
how good will be the final goal of ill, if not
in this life, in another.
‘ Euripide excelle surtout 4 troubler,’ says
M. Jules Girard, and Robert Browning testi-
fies to the same when he speaks of the many
unsuspected temptations to faith that will
trouble the most comfortable atheism, just
as doubts will trouble faith :—
‘And now what are we? unbelievers both.
Where’s the gain? .
Just when we are ey theres! a ἘΠΕ:
touch,
A fancy from a flower-bell, some one’s
death,
A chorus-ending from Euripides.—
And that’s enough for fifty hopes and fears
As old and new at once, as nature’s self,
To rap and knock and enter in our soul,
Take hands and dance there, a fantastic
ring
Round the ancient idol, on his base again,
The grand Perhaps.’
Was it this that made Aristotle call him
the most tragic of the poets, that strong
appeal for ‘more light’ which Clement of
Alexandria has preserved from some lost
play ἴ--
πέμψον μὲν φῶς ψυχᾶς ἀνέρων
τοῖς βουλομένοις ἄθλους προμαθεῖν
πόθεν ἔβλαστον, τίς ῥίζα κακῶν,
τίνα δεῖ μακάρων ἐκθυσαμένοις
εὑρεῖν μόχθων ἀνάπαυλαν.
In spite of Aristophanes and of those who
condemned Socrates, the sceptism of Euri-
pides and of Socrates was latent in the
minds of most of their contemporaries. The
well-known stories of the popularity of
Euripides are sufficient proof that his voice
awoke an answering note in the hearts of his
audience. Sophocles himself, says Aristotle,
said that while he described things as they
ought to be, Euripides described things as
they are ; and Sophocles no doubt thought
that the Athenians ought to have worshipped
with quiet confidence and submission the
gods of their fathers, leaving the mysteries
THE CLASSICAL REVIEW. 93
of life and death in their hands. But
in very deed they were, as Euripides divined
them, deep in their hearts troubled and in
doubt. Their very wrath with those who
gave voice to their trouble was only the last
effort of religious conscientiousness to stifle
the over-mastering impulse within. Those
who condemned Socrates of speaking blas-
phemously were themselves uttering the
cynical ‘What is truth?’
lt follows that the allusions to the after-
state in Euripides, though frequent, are
vague and conjectural. Sometimes, for the
purposes of his play, he makes use of the
traditional imagery and legend ; sometimes
he openly rejects them as μῦθοι. He does
not introduce apparitions of the dead or
demon-phantoms. The Erinnyes seen by
his Orestes are evidently the products of
a distempered brain, and the eidolon of
Polydorus, which appears in order to speak
the prologue to the Hecuba, is only a piece
of stage-machinery. We may take, as an
example of his treatment, the play of
Alcestis.
Alcestis—The idea of redemption from
death by the voluntary substitution of
another person is not without parallel else-
where in Greek legend. There is, for
example, the sharing of immortality and
death between the Dioscuri and the re-
demption of Prometheus from Tartarus by
Cheiron in the Prometheus Unbound. But
it is hardly necessary to say that Euripides
did not believe, any more than the Hebrew
poet, that man could release his brother’s
soul from death, or make agreement unto
God for him. He treats the story as we
should, as a beautiful fable, illustrative of
wifely devotion and love that conquers
death, and as a convenient vehicle for what
is to us the less interesting motive, though
the more important one in intention—that
an unselfish regard for the duties of hospi-
tality and of the obligations to noble self-
restraint is not in time without its own
rewards. Thanatos is a purely allegorical
figure, not exactly answering to any of the
Chthonian deities, just as the restoration to
life of Alcestis is treated symbolically by
the contest between Thanatos and Heracles.
Again the dying wife sees in fancy all
that Greek imagination had figured of the
under-world : the two-oared bark ; the ferry-
man of the dead, his hand upon the oar ;
winged Hades himself, with burning eyes
that frown darkly at her under his beetling
brows. But the poet is only putting into
her mouth the expressions appropriate to
the long-past age to which his heroine may
be supposed to belong: they did not repre-
sent any belief of his own or his audience.
Aristophanes might parody the same old
fables without fear of censure in his Frogs.
Euripides has distinctly said that we know
not the meaning of this glittering bauble
life, much less of what comes after. ‘Sleep,
rest, and peace are all we know of death’
might translate many a passage in his
plays.
The poet does not permit Alcestis herself
any soothing prevision of recompense for
her heroism ; but three times the Chorus—
who on the whole may be supposed to speak
for Euripides—venture a conjectural hope
that her lot in Hades may be made worthy
her deserts :—
ΔῊΝ. εἰ δέ τι κἀκεῖ
πλέον ἐστ᾽ ἀγαθοῖς, τούτων μετέχουσ᾽
ἽΑιδου νύμφῃ παρεδρεύοις.
Nay, they half canonize Alcestis, bidding
Admetus to honour her tomb, not with the
customary melancholy rites, but as the altar
of a goddess :—
4 Ν + =) Be) Ν
αὕὗτα ποτὲ προύθαν᾽ ἀνδρὸς
~ >
νῦν δ᾽ ἐστὶ μάκαιρα δαίμων.
nA? ὦ 7 > Ν if
χαῖρ, ὦ ποτνί᾽, εὖ δὲ δοίης.
And what of consolation for Admetus ?
That she will prepare a mansion where he
may join her :—
> aA 35 ὦ
ἀλλ᾽ οὖν ἐκεῖσε προσδόκα μ᾽ ὅταν θάνω
r 3 2
καὶ δῶμ᾽ ἑτοίμαζ᾽, ὡς συνοικήσουσά μοι,---
a reunion which his own funeral shall
symbolize, laying him to rest beside her :—
>. “ 3 - Τὰ ΘΠ , “ὃ
ἐν ταῖσιν αὐταῖς γάρ μ᾽ ἐπισκήψω κέδροις
~~ -ὰ 4
σοὶ τούσδε θεῖναι πλευρά τ᾽ ἐκτεῖναι πέλας
πλευροῖσι τοῖς σοῖς.
In the meantime, while Admetus remains
to drag out a weary existence, παρεὶς τὸ
μόρσιμον, ‘her sorrow shall in no wise touch
any more; she has ceased from troubles,
fair-famed for ever.’
This is the grain of comfort, limited, alas!
by many an if. Euripides however, we are
fain to believe, leant like Socrates towards
hope. The end presented itself to him, as
to Socrates, as offering two alternatives,
and the apparent contradictions of some
of his utterances are due to this. A con-
ception of the Deity as absolute good,
consciousness of capabilities only half re-
alized, of the far-away mysteriousness of
God, τὸν πάνθ᾽ ὁρῶντα καὐτὸν οὐχ ὁρώμενον,
of man himself —‘animus ut Euripides
dicere audet deus est,’ says Cicero (Zusc.
94 THE CLASSICAL REVIEW.
I. 26)—all this at times inclined him to
the thought that death may after all be
the better part of existence. But again the
weariness and sorrows of this life, its con-
tradictions and its failures, pressed in with
stern insistence, and it was enough that
death should end the pain and wretchedness
in sleep dreamless and eternal :—
/ Ν 3 ’ / i
θάνατος yap ἀνθρώποισι νεικέων τέλος
3, ’, Ν AQ? > \ a 5 A
ἔχει: τί yap τοῦδ᾽ ἐστὶ μεῖζον ἐν βροτοῖς ;
τίς γὰρ πετραῖον σκόπελον οὐτάζων δορὶ
58 / 4 3.95 ΄ 4
ὀδύναισι δώσει ; Tis δ᾽ ἀτιμάζων νέκυς,
εἰ μηδὲν αἰσθάνοιτο τῶν παθημάτων ;
Frag. of an Antigone.
In this mood he said in the Cresphontes
that we should not, 85. we do, celebrate
birth with thanksgiving and death with
mourning. Rather should we mourn when
a child is born into this world, pondering
on all the sorrows of our human life; but
when another has ended all these cruel
hardships in death, then with praise and
joyfulness should his friends escort him
forth upon his way (Cic. Zusc. I. 48). At
another time, in his bitterness and disgust,
he is ready to be rid of humanity alto-
gether :—
Ν ἴω S A Ν Ν Lal > Ν
τοὺς ζῶντας εὖ δρᾶν: κατθανὼν δὲ πᾶς ἀνὴρ
γῆ καὶ σκιά: τὸ μηδὲν εἰς οὐδὲν ῥέπει---
Dust and shadow, potential nothingness in
life, actual nothingness in death. Our
enemies can harm us no more, our friends
forget us eqrally :—
τύμβῳ yap οὐδεὶς πιστὸς ἀνθρώπων pitos—
Frag. incert. Fab.
just as we forget the dead in enjoying the
inheritance they have left us :—
TO yap ἔχειν πλέον κρατεῖ
τῆς εὐσεβείας: ἡ δ᾽ ἐν ὀφθαλμοῖς χάρις
5 / ? Ψ 5 / 3
ἀπόλωλ᾽, ὅταν τις ἐκ δόμων ἔλθῃ.
Ibid.
Only their works remain to keep green the
memory of the just; this immortality at
least is theirs :—
ἀρετὴ δὲ κἂν θάνῃ τις οὐκ ἀπόλλυται,
- > 5 a
ζῇ δ᾽ οὐκετ᾽ ὄντος σῶματος" κακοῖσι δὲ
A “ἅμ / ae Ν ’
ἅπαντα φροῦδα συνθάνονθ᾽ ὑπὸ χθονός.
Frag.
Still, in his gloomiest mood Euripides would
have us remember that if we know not good,
neither do we know evil—our ignorance is
absolute. Nothing remains for us but to
cultivate a philosophic spirit in face of this
inevitable mystery. Life is ours to make
the best of it, whatever comes after, and if
death is part of the course of nature, a
harvesting of ripe fruit, there is at least
that presumption that it is somehow for
the best :—
4 Ν > ν᾽ Ὁ 3 a a
ἔφυ μὲν οὐδεὶς ὅστις οὐ πονεῖ βροτῶν"
θάπτει τε τέκνα χἄτερ᾽ αὖ κτᾶται νέα,
3 / / Ἂς sO?» A
αὐτός τε θνήσκει. καὶ τάδ᾽ ἄχθονται βροτοὶ
“ an 3
εἰς γῆν φέροντες γῆν" ἀναγκαίως δ᾽ ἔχει
βίον θέριζειν ὥστε κάρπιμον στάχυν,
καὶ τὸν μὲν εἶναι, τὸν δὲ μή: τί ταῦτα δεῖ
στένειν ἅπερ δεῖ κατὰ φύσιν διεκπερᾶν ;
δεινὸν γὰρ οὐδὲν τῶν ἀναγκαίων βροτοῖς.
Frag. Hypsipyle.
This is what the tragedians leave with
us: Aeschylus, his trumpet-note of judg-
ment ; Sophocles, his tenderer words of rest
and reunion; Euripides, his troubled ques-
tioning, his sceptic silence, his half-con-
temptuous calm. Between the birth of
Aeschylus and the death of Euripides most
of the great things of the Athenian empire
are comprised, and when Euripides died the
turmoil and distress of the later years of
the Peloponnesian war was fast closing in
the political horizon, and thrusting men’s
thoughts back upon themselves. And as
the soldiers of Plataea could not but feel
the value of life and action, and the citizen
of the Athens of Pericles and Pheidias could
not but feel the value of human _ per-
sonality (πολλὰ τὰ δεινὰ κοὐδὲν ἀνθρώπου
δεινότερον πέλει), So the poet who had out-
lived the disasters of Syracuse, and had
heard how the Odes that had been written
to delight the gay leisure of the free proud
sons of Athens had come to be the solace
and alleviation of slavery and imprisonment,
could not but have felt something of the
vanity of life and of human achievement,
and have asked more eagerly than his
compeers, What lies beyond ?
There is at the British Museum a beautiful
Graeco-Roman bronze head of Sleep or her
twin Death. The parted hair, the dove-like
wings over the delicate ears, the eyelids that
droop low over the languid eyes, the small
straight nose, the close quiet mouth and
firm chin—all breathe an indescribable air
of gravity and reticence and calm, which
seem to say that this is the last and
crowning experience of all, that this is the
secret of all secrets, which when we have
fathomed, we have fathomed all. And we
stand before her, each with his separate
THE CLASSICAL REVIEW. 95
puzzle, his own vexed thoughts, his bafiled
questionings, his blank misgivings. She
could give each his answer, but she will
not speak for all our entreaties—not yet.
There in bronze is the Greek thought of
Death. Volumes about it could tell us no
more.
Maup M. Danre..
POLITICAL ALLUSIONS IN THE SUPPLICES OF EURIPIDES.
Or the three great tragic poets of Athens
Euripides most concerns himself with the
political questions of the time. The Hercules
Furens, the Heraclidae, the [on all show this
tendency to some extent, but of all Euripides’
plays the Swpplices is undoubtedly the most
political. The disaster at Delium in 424 B.c.
and the refusal of the Thebans contrary to
Greek religious custom to give up the bodies
of the dead manifestly suggested the theme
of the play. Unfortunately no ancient
authority gives us the date of its production,
and the opinions of modern writers vary.
The majority place it somewhere between
422 and 417 B.c.: some as late as 410.
Hermann suggested that as the play so
manifestly supports the Argive alliance it
was produced before the Argive ambassadors
in 420. Various passages in the play have
been understood by various writers to apply
to eminent Athenian politicians of the day.
Wilamowitz-Mollendorf and others consider
it practically a party pamphlet in the
interest of Alcibiades. But the passages
which have been referred to are too isolated
to clearly prove the poet’s political intention.
The most important and most certain proof
these writers have one and all passed over.
It is contained in the long panegyric of
Adrastus delivered over the bodies of the
fallen chiefs (857 ff.).
The first of the heroes whom Adrastus
describes is Capaneus who had died before
the walls of Thebes ‘by the visitation of
God.’ Aeschylus (Septem ὁ. Thebas 422)
describes how Capaneus declared that he
would take the city and defied Heaven to
prevent him. Euripides repeats the passage
almost verbatim (498) ὥμοσεν πόλιν | πέρσειν,
θεοῦ θέλοντος ἦν τε μὴ θέλῃς. To the Greek
mind, the death of Capaneus by lightning
was the natural result of such presumptuous
sin. It is consequently with some astonish-
ment we see that in Adrastus’ account of
the chieftains, although the original story of
the death of Capaneus by lightning is kept,
still his character is represented as all that
is good and amiable. Of the haughty Ca-
paneus braving in his pride the wrath of
Heaven we read (862) φρόνημα δὲ | οὐδέν τι
μεῖζον εἶχεν ἢ πένης ἀνήρ and again (869)
ἄψευδες ἦθος, εὐπροσήγορον στόμα.
The editors explain this glaring inconsist-
ency by the plea that the words are in the
mouth of a friend, and that the cause of
Adrastus was bound up with his deceased
allies. But this did not prevent Adrastus
saying before (160) νέων yap ἀνδρῶν θόρυβος
ἐξέπλησσέ pe. Nor is Theseus anywhere
in the play inclined to palliate the conduct of
the Seven although he feels himself bound in
the interests of Greek religion to demand
the bodies for the Suppliants at the altar of
Demeter.
It might also be urged that the poet
desired to lend some novelty to his treatment
of the subject which, as Plutarch (Theseus ¢.
29) tells us, had already been handled by
Aeschylus in his Lleusinians. It is not
without reason that Aristophanes makes
Euripides the claimant of the throne of
Aeschylus in the lower world. Not a few
times he goes out of his way to show his
rivalry with Aeschylus. In this very play
the speech of Theseus (195 ff.) is obviously
modelled on and meant to rival the speech
of Prometheus in Aeschylus’ play of that
name (P. V. 442 ff.). We learn from
Plutarch that in the Hleusinians Aeschylus
made the bodies be given up to Theseus after
peaceable negotiation and without the battle
which Euripides introduces. But as Euripi-
des followed Aeschylus earlier in the play in
his account of Capaneus, and as there is no
reason to doubt the authenticity of that pas-
sage, there must be some other reason for his
inconsistency with himself. That reason can
only be that he wished to draw under the
guise of the Argive chieftains the political
chieftains of Athens.
The only editor, so far as I know, who
has seen this is Musgrave. , His words are
...puto poetam in ducibus Argivis laudandis
nobilium quorundam Atheniensium vitam et
mores effinaisse. Quare non est adeo mirandum
si receptam de veteribus ils opinionem non
anxie ubique secutus est. But Musgrave
made no attempt to work out the matter in
90 THE CLASSICAL REVIEW.
detail. When once, however, the speech is
looked at from this point of view it is com-
paratively easy to see who are the characters
depicted. No doubt the youthful citizens
to whom Adrastus is requested by Theseus
to tell the names and characters of his
comrades recognised with little trouble the
politicians whom the Argive leaders were
intended to represent.
(i.) The irreligious Capaneus is metamor-
phosed into the god-fearing Nikias, ᾧ Bios
μὲν ἣν πολὺς | ἥκιστα δ᾽ ὄλβῳ γαῦρος ἣν (861).
The wealth of Nikias was proverbial not
only among his own contemporaries (Lysias,
xix. 47, says his property was estimated at a
hundred talents) but also in much later
times (Athenaeus, vi. 2726 τῶν Ἑλλήνων
ζάπλουτος Νίκιας). Cp. also Plut. Wik. xv. 2
τοῦ δὲ Νικίου καὶ διὰ τἄλλα μέγας ἣν καὶ διὰ
τὸν πλοῦτον καὶ διὰ τὴν δόξαν ὃ ὄγκος.
Plutarch’s character of Nikias is somewhat
inconsistent with itself. As it is stated in
il. 3 it agrees admirably with the portrait
here—xat γὰρ οὐκ ἣν αὐστηρὸν οὐδ᾽ ἐπαχθὲς
ἄγαν αὐτοῦ τὸ σεμνόν, ἀλλ᾽ εὐλαβείᾳ τινὶ
μεμιγμένον αὐτῷ τῷ δεδιέναι δοκοῦντι τοὺς πολλοὺς
δημαγωγοῦν. A poet who, as we shall see, was
a more or less ardent sympathiser with the
aims of Nikias, was not likely to dwell upon
his weak points nor indeed, as we can see
from his unhappy future history, were these
so evident to his contemporaries as they have
been to posterity. In any case it is hard to
reconcile the passage given above with the
statement (v. 4) that he was φίλοις οὐ προση-
νὴς οὐδ᾽ ἡδύς K.T.A. Or With xi. 2 Νικίαν ὃ τε
πλοῦτος ἐπίφθονον ἐποίει καὶ μάλιστα τῆς διαίτης
τὸ μὴ φιλάνθρωπον μηδὲ δημοτικὸν ἀλλ᾽ ἄμικτον
καὶ ὀλιγαρχικὸν ἀλλόκοτον ἐδόκε. Nor are
these latter passages reconcilable with the
statement (11. 2) τὸν δῆμον εἶχεν εὔνουν καὶ συμ-
φιλοτιμούμενον---ὃ, statement which is indis-
putably true. Of no one could it be said
more truly than of Nikias ἄκραντον οὐδὲν
οὔτ᾽ ἐς οἰκέτας ἔχων οὔτ᾽ ἐς πολίτας (870).
(ii.) The Athenian who is represented by
Eteoclus is equally easy to discover. The
characteristic on which Euripides dwells is
the poverty of the hero and his stern inde-
pendence. One of the leading men of the time
had precisely the same characteristics.
Plutarch (Nikias, xv. 1) tells us that Lama-
chus was a brave and honest man (jv μὲν
ἀνδρωδὴς καὶ δίκαιος ἀνήρ) but so poor that he
had always in his accounts to debit to the
Athenians a small sum for clothes and shoes.
We are told in the life of Alcibiades by the
same writer that owing to his poverty he
had not the influence he deserved (Alcib. 21).
Aristophanes in Lamachus’ life-time loved to
represent him as a swashbuckler and a sort
of Captain Bobadil, but generously sang his
praises after he had met his death in the
service of his country (Zhesmoph. 841, Ranae,
1039).
(111.) Hippomedon, the chief who ‘scorns
delights and lives laborious days’ in his
desire to serve his country, seems to me to
be Demosthenes, perhaps the greatest of
all Athenian generals. Unfortunately no
full account of him is preserved to us. He
seems to have been a man of war from his
youth and to have taken no direct interest
in politics. In Aristophanes he appears only
in the Hquites as οἰκέτης A. Thucydides
seems to think he needs no introduction to
his readers, at any rate no special attention is
drawn to his appearance on the political
stage.
(iv.) The antitype of the next hero—
Parthenopaeus the son of Atalante—is plain
to all eyes. To only one Athenian of
Euripides’ time could the words Παρθενοπαῖος,
εἶδος ἐξοχώτατος and πολλοὺς δ᾽ ἐραστάς κἀπὸ
θηλειῶν *6 παῖς" 1 ἔχων ἐφρούρει μηδὲν ἐξαμαρ-
τάνειν apply—to Alcibiades. The verses 899,
900 coming at the end of the character
seem thrown in as an afterthought of the
poet in order to make identification absolutely
certain. How true they were to Alcibiades’
character may be seen by comparing Xen.
Memorab. i. 2, 24 and Thue. vi. 15. Aleci-
biades was just entering on his political
career ; there had been an attempt to snub
him, but he had successfully revenged that
and had suddenly found himself the author
of a policy at least partially adopted by
Athens. He would have been considered
very young for a politician in any other
state, says Thucydides (v. 43, 2). Are the
lines λυπηρὸς οὐκ ἣν οὐδ᾽ ἐπίφθονος πόλει | οὐδ᾽
ἐξεριστὴς τῶν λόγων κιτιλ. (all of which
Alcibiades already was or soon became) a
gentle hint from the poet of conduct which
his well-wishers and his country might think
it well he should amend ?
(v.) In the account of Tydeus there exists
a very serious textual difficulty, Something
has been lost after δεινὸς σοφιστής, or some-
thing has been interpolated ; perhaps there
have been both loss and interpolation. But
the lines that remain give an excellent clue
to the personage meant by the poet. The
historical character must be the beauw-idéal
of the courageous warrior; he must bea man
of deeds not arguments, οὐκ ἐν λόγοις ἣν λαμ-
πρὸς, ἀλλ᾽ ἐν ἀσπίδι δεινὸς σοφιστής. Fortu-
nately we have an account of such an
1 Emendation of Wilamowitz (Hermes, xiv. 181)
for the unintelligible ὅσας of MSS.
THE CLASSICAL REVIEW. om
Athenian preserved to us by Plato. It is
Laches. Τὸ both Euripides and Plato one
person has acted as the exemplar of εὐανδρία.
Plato sketches exactly the same bluff rude
soldier as Euripides. Laches knows his own
defects, as he tells us in his account of
Socrates (188 E), Σωκράτους δ᾽ ἐγὼ τῶν μὲν
λόγων οὐκ ἔμπειρός εἰμι, ἀλλὰ πρότερον, ὡς
ἔοικε, τῶν ἔργων ἐπειράθην (cp. also 194 ΑἹ:
πρότερον refers to the battle οἵ Delium (181
B), at which Socrates showed his superiority
in bravery even to Laches himself with whom
he retired from the field (Plato Symposium
221 A).
This completes the list of heroes who have
a funeral oration pronounced over them.
Amphiaraus had been swallowed up by the
earth and Polyneices’ corpse had been left at
Thebes. Hence they are passed over with
three lines each and these from Theseus not
Adrastus. Nor is there any attempt to
delineate their characters.
The list of persons mentioned strongly
supports Hermann’s date for the play. Of
the five Athenians described—Nikias, Lama-
chus, Demosthenes, Alcibiades, Laches—all
except Alcibiades had sworn to the treaty of
Nikias in the previous year (Thue. v. 19) and
they are indisputably the best known person-
ages among the Athenian deputies who
were present on that occasion. It was since
that treaty that Alcibiades’ star had risen.
Before that treaty he had been regarded by
the Spartans as a person of no consequence ;
he was now at the head of the most influen-
tial’party in Athens. No wonder then that,
with those traits of character which had
already shown themselves in him, the poet
should warn him against his besetting sins.
When this, the most important passage
politically in the play, has been set in its
proper light, the political allusions scattered
elsewhere throughout the play appear in their
correct relations, and the whole purpose of
the drama is made clearer. Euripides’ aim
is universal peace. The poet feels that men
ought to heal their differences by milder
remedies than the sword— κενοὶ βροτῶν |
οἱ τόξον ἐντείνοντες ὡς καιροῦ πέρα, | καὶ πρὸς
δίκης γε πολλὰ πάσχοντες κακά, | φίλοις μὲν οὐ
πείθεσθε, τοῖς δὲ πράγμασι, | πόλεις τ΄, ἔχουσαι
διὰ λόγου κάμψαι κακά, | φόνῳ καθαιρεῖσθ᾽. οὐ
λόγῳ, τὰ πράγματα (144 ff.). Let there be
a treaty with Argos by all means, but that
need not entail a war with Sparta. This
was, as we learn from Thucydides, the gene-
ral feelin of the Athenians on the subject
(Thue. v. 48, 1 αἱ μὲν σπονδκὶ καὶ at ξυμμαχίαι
οὕτως ἐγένοντο καὶ at τῶν Λακεδαιμονίων καὶ
᾿Αθηναίων οὐκ ἀπείρηντο τούτου ἕνεκα οὐδ᾽ ὑφ᾽
ΝΟ. XXXI. VOL. IV.
ἑτέρων). This is supported also by the care-
ful avoidance of anything which might
offend Sparta unnecessarily. The only direct
reference of any kind (187) Σπαρτὴ μὲν ὠμὴ
καὶ πεποίκιλται τρόπους iS in a passage part
of which is certainly spurious. Dindorf
brackets vv. 180 to 192. All editors are
agreed that vv. 180—183 are spurious.
Kirchhoff and Nauck bracket also vv. 190
—192, which some have taken for a pane-
gyric of Alcibiades. The poet’s feelings
and those of his audience must have been
different on this occasion to those they
entertained at the production of the Andro-
mache. The passage in the Andromache (445
—463) beginning ὦ πᾶσιν ἀνθρώποισιν ἔχθιστοι
βροτῶν | Ξπάρτης ἔνοικοι gives a list of the
failings of Sparta which reminds one of the
views expressed by the Athenians in the
Melian dialogue. Even if the solitary verse
in this play be kept it might be looked upon
as a sight hint of the irritation the Atheni-
ans were feeling because Sparta had
not coerced her allies into peace and had
not compelled the Thebans to surrender
Panactum.
But if Euripides’ desire was for peace he
could not have been a thorough supporter of
Alcibiades. Here therefore ἢ must dissent
from the conclusions drawn by Wilamowitz-
Mollendorf (Analecta Euripidea p. 179).1 He
there points out that Aristophanes, however
merciless to Euripides’ earlier and later plays,
heaps none of his contumely on the cycle of
political plays—the Hercules Furens, Heracli-
dae, Cresphontes, Erechtheus, Supplices, Ion—
which he assigns with great probability to
the middle of the poet’s career. He therefore
argues that the reason for this must be that
at this period both poets were fighting on the
same side. That side he considers to have
been the side of Alcibiades. There is however
no proof that either poet was a partisan of
Alcibiades. Aristophanes whose every de-
sire is peace at any price could hardly be so.
Neither could Euripides with the sentiments
he develops in this play. The party which
he favours is very clearly indicated (244 f.)
τριῶν δὲ μοιρῶν ἡ ᾽ν μέσῳ σῴζει πόλεις κόσμον
1 Since this paper was in the hands of the editor
I have had an opportunity of glancing at the first
volume of W.M.’s new work—Herakles. It is pleas-
ing to find so eminent an EKuripidean scholar also ex-
pressing the opinion (p. 13) that the object of this
play was ‘frieden mit Sparta, aber anschluss an Argos
zu suchen’; but, as will be seen, I cannot accept his
further conclusion that, at the time of the production
of this play at any rate, Euripides hoped to find in
Alcibiades a greater Pericles. This conclusion is
founded on vv. 190—192 which, as already remarked,
are regarded as spurious by several of the best
editors.
H
98 THE CLASSICAL REVIEW.
φυλάσσουσ᾽ ὅντιν’ ἂν τάξῃ πόλις. OF the
democratic side which Alcibiades was then
favouring he uses very plain words (243)
γλώσσαις πονηρῶν προστατῶν φηλούμενοι, and
sketches the demagogue in the darkest colours
(412 ff.). Huripides will have none of the
young party who πολέμους αὐξάνουσ᾽ ἄνευ
δίκης, | φθείροντες ἀστούς, ὃ μὲν ὅπως στρατη-
λατῇ, ὃ 8 ὡς ὑβρίζῃ, δύναμιν ἐς χεῖρας
λαβών, [ ἄλλος δὲ κέρδους οὕνεκ᾽ οὐκ ἀποσκο-
πῶν | τὸ πλῆθος εἴ τι βλάπτεται πάσχων τάδε
(233 ff.), words which were as true applied to
those who wished to upset the peace of Nikias
in 420 as in the mouth of Nikias himself
before the Sicilian expedition (Thuc. vi.
ΤΠ: 92).
It is perhaps worth observing how careful
Euripides is to make his audience see that he
is not really describing a mythical scene, but
one in which the majority of themselves
have taken part. Little illusion as to the
meaning of the play could be left after the
reply of Theseus to the Theban herald (404)
οὐ yap ἄρχεται ἑνὸς πρὸς ἀνδρός, ἀλλ᾽ ἐλευθέρα
πόλις ἃ statement which is somewhat out of
place in the mouth of even the most consti-
tutional of ancient monarchs. The play is
permeated with references to other events of
the day. The words of Theseus (532)
ὅθεν δ᾽ ἕκαστον ἐς τὸ φῶς ἀφίκετο | ἐνταῦθ᾽
ἀπῆλθε, πνεῦμα μὲν πρὸς αἰθέρα, | τὸ σῶμα δ᾽ ἐς
γῆν are but a paraphrase of the inscription
on those who fell at Potidaea (Hicks, /nscript.
p. 60) αἰθὴρ μὲν ψυχὰς ὑπεδέξατο, σώ ματα δὲ
χθών] | τῶνδε κιτιλ. The lines 847 ff. seem
a distant reminiscence of the reply of the
Spartan captured in Sphacteria to the
Athenian ally who asked him εἰ οἱ τεθνεῶτες
αὐτῶν καλοὶ κἀγαθοί (Thue. iv. 40). The
course of the battle was obviously meant to
remind the audience of Dehum. At the
battle of Delium the right wing of the
Athenians drove back the Boeotians while
their own left wing which was opposed to
the Thebans was defeated (Thuc. iv. 96, 3,
4). Compare with this (Suppl. 703) λόχος δ᾽
ὀδόντων ὄφεος ἐξηνδρωμένος | δεινὸς παλαιστὴς
ἣν" ἔκλινε γὰρ κέρας | TO λαιὸν ἡμῶν" δεξιοῦ δ᾽
ἡσσώμενον | φεύγει τὸ κείνων, ἣν δ᾽ ἀγὼν ἰσόρ-
ροπος. Valuable light is tlrown on the λόχος
ὀδόντων ὄφεος by a passage in Diodorus (xii.
70) which tells us that in the front of the
Theban force there fought a chosen band of
three hundred, οἱ wap’ ἐκείνοις ἡνίοχοι καὶ παρα-
Barat καλούμενοι. The names are specially
interesting because they show that the band
was really a survival from a time when the
heroic methods of fighting were still in vogue
and might therefore be attributed by
Euripides without difficulty to the heroic
period,
Finally, when we remember that Amphia-
raus had fled from before the walls of Thebes
through the district of Oropus where the
Athenians sought refuge after Delium, and
that according to the legend he disappeared
at Psaphis (Strabo, 6 399, Philostratus Jm-
agines i, 26, 3), we see what opportunities
Euripides had for connecting the battle of
Delium with the story of the Seven and how
cleverly he availed himself of these opportu-
nities.
It has been made an objection to the play
that it has no action. That is true. But
the Athenian absorbed in reviewing the
history of his own time unrolled before him
with only a change of names would not have
felt this any more than he would have felt
it in the Persae. The play is—as the frag-
ment of the ὑπόθεσις describes it—an ἐγκώμιον,
᾿Αθηνῶν and a much more skilful one than
the Persae. It was easy for Aeschylus to
make his panegyric after a great triumph ; it
required much greater skill to do so after a
great defeat. It is not in either case the
highest form of dramatic art. But Euripi-
des desired not only to praise Athens. He
had a policy to support and a lesson to convey
to his countrymen and, such being his aim, he
could not have taken better means to attain
it.
P. GiLEs.
THE PLOT OF THE AGAMEMNON.
Dr. Verratt’s brilliant and ingenious
introduction to the Agamemnon opens up a
question which it will take time to decide
upon. Having been much discontented for
a long time with the plot of the play as we
have all been taught it, I at any rate am
glad to have something set before me which
satisfies such modest demands as_ the
aesthetic feeling of every reader makes with
regard to the construction, or the skeleton,
of a tragedy. It is in the hope that the
following considerations may do something to
further elucidate this difficult matter that I
submit them to the reader.
THE CLASSICAL REVIEW. 99
Dr. Verrall lays great stress, as we all
know, on the beacons, a feature of the play
before inexplicable. Now it has never, I
think, been pointed out in this connexion
that in another story of the voyage home
from Troy beacons play an important and
disastrous part. Nauplius, the father of
Palamedes, in revenge for his son’s death,
lighted false beacons and so caused the
returning fleet to suffer shipwreck. Accor-
ding again to the Scholiast on Lycophron’s
Alexandra, 1093, Ναύπλιος διατρίβων ἐν
Ἑλλάδι παρεσκεύασε μοιχευθῆναι τὰς γυναῖκας
τῶν ἐν τῇ Τροίᾳ: καὶ Κλυταιμνήστραν μὲν
συνέμιξεν Αἰγίσθῳ κιτιλ So then in one story
Nauplius and his beacons (ep. Lycophron
1096) are connected with the adultery of
Aegisthus and Clytemnestra to bring about
the ruin of the fleet and of Agamemnon ; in
the other story, that of Aeschylus, or of Dr.
Verrall if you lke, beacons and the adul-
terous pair are connected for the ruin of
Agamemnon. This is hardly an accidental
coincidence.
Again, Aeschylus must have known the
story of Nauphus. For Nauplius was one
of the characters of the Νόστοι. πρὶν δὲ
τελευτῆσαι ἔγημεν [ὃ Ναύπλιος), ὡς μὲν ot
τραγικοὶ λέγουσι, Κλυμένην, ὡς δὲ ὁ τοὺς
νόστους γράψας, Φιλύραν. Apollodorus Lidl.
I. 2 (page 40 Teubner ed.). And the story
of the shipwreck was told in the Νόστοι.
εἶθ᾽ ὃ περὶ tas Kadnpidas πέτρας δηλοῦται
χειμών, Says the Chrestomathia οἵ Proclus, in
the epitome of the Νόστοι. It is no great
assumption then that the beacon of Nauplius
was a familiar story in the time of Aeschylus;
at any rate it is more likely that he adapted
this beacon story to his own ends than that
there were two independent stories about
beacons.
Can the beacon, one may ask, be doing
double duty ?° Is it both a signal to Clytem-
nestra to be ready, and a false fire to destroy
the Greek fleet? Aeschylus does not suggest
this and it seems improbable, though it
would give extra point to πῦρ καὶ θάλασσα in
Agam. 656. Rather he prefers to make the
beacon help in destroying Agamemnon in
another way than in the old story, he or the
authority he follows. He keeps the signal
because it was in the legend, but drops
Nauplius because he is not wanted and is
better away.
If Dr. Verrall is right the conspirators
had information from Troy. Why then
watch for a year? They need not have
begun watching till they heard of the capture
of Troy. The inconsistency is that (1) they
had news continually, (2) they fear they may
be taken unawares. This comes of the epic
story being transplanted. Homer naturally
supposes they have no news, and he sets a
σκοπός. Aeschylus naturally supposes they
have news but keeps the σκοπός all the same,
and leaves the inconsistency to look after
itself as Sophocles does with much worse
ones in Oedipus Rex. If he had troubled
himself about it, he would have made the
φύλαξ only have been looking out for a week
or two, but it is no wonder that he prefers the
Homeric year.
If we are to fill up the story in all its
details we must suppose that watch was kept
by day as well as by night, for a fleet is
more likely to return by day of the two.
But if the story said they returned by night,
why should the poet trouble himself further ?
Dr. Verrall justly lays stress on and
defends line 1644 :
> IN > / 3 Ν Ν ta
οὐκ αὐτὸς nvapiles, ἀλλὰ σὺν γυνή.
In further defence it may be observed that
this line is a reminiscence of Homer λ. 409:
ἔκτα σὺν οὐλομένῃ ἀλόχῳ. In both the point
is the same, that Aegisthus devised the plot
with the help of Clytemnestra.
The great difficulty in Dr. Verrall’s view
seems to be the φύλαξ. The Homeric watch-
man knows perfectly what he is about ; why
does the Aeschylean play a part he does not
understand ? It is due, I think, to the Attic
irony, which is powerfully developed in his
joy at what he supposes the sign of his
master’s success while it is really the sign of
his ruin, and in his being the unconscious
instrument, like Emilia, of what he least
wishes.
The great speech in which the beacons are
described may well have had a special
interest for the audience, which would ex-
cuse its falsity, if a λαμπαδηφορία were to
take place that evening after the tragedies
were over. Clytemnestra herself speaks of
it as a λαμπαδηφορία.
ARTHUR Puatr.
H 2
100
THE CLASSICAL REVIEW.
ON THE CONSTRUCTION OF CLAUSES FOLLOWING EXPRESSIONS OF
EXPECTATION IN GREEK.
In a note on Babrius ix, 2, discussing the
constructions following expressions οἵ
‘hoping’ in Greek, Mr. Rutherford has done
good service in removing one of the few
tares which have cropped up in the wheat of
Mr. Goodwin’s Moods and Tenses, itself
indeed a καλλίπταις ἀμητὸς ἐλπίδων πληρής. Mr.
Rutherford there points the distinction
between examples like ὄψον ἐλπίσας ἥξειν and
ἐλπίς ἐστιν ἥκειν. In the former a future form
of the inf. is required, because the inf. is
oblique and so stands for a fut. indic. of the
direct ; but in the latter, where the inf. is
substantival, the pres. or aor. (without ἀν)
may stand—indeed the full form ἐλπίς ἐστι
τοῦ ἥκειν often occurs, as Thue. 11. 56, εἰς
ἐλπίδα ἦλθον τοῦ ἑλεῖν. It appears however
that Mr. Rutherford has overlooked another
form, in which, though the governing phrase
is verbal (ἐλπίζω, etc., not substantival ἐλπίς
ἐστι etc.), the governed infinitive may still
be substantival, and therefore does not
necessarily take the mark of future time.
We refer to those cases in which the inf.
has no subject expressed. There is the same
difference between ἐλπίζω σε ἥξειν, ‘I hope
that you will come,’ and ἐλπίζω ἥκειν, ‘I hope
to come,’ as there is between φησί σε ἐλθεῖν,
‘he says that you came,’ and βούλεται ἐλθεῖν,
‘he wishes to come.’ All this may seem
very elementary, and indeed Mr. Goodwin
has already stated quite clearly (p. 14) ‘the
Greek makes no more distinction than the
English between ἐλπίζει τοῦτο ποιεῖν, ‘he
hopes to do this,” and ἐλπίζει τοῦτο ποιήσειν,
“he hopes that he shall do this.”’ What,
however, he does not appear to point out
with sufficient emphasis is, that if the infin.
had a subject expressed, only the latter of
these forms would be admissible in Attic:
when the inf. has no subject, either form
may stand. Hence, we venture to think,
his remarks on p. 33 (δ 23, 2) lack precision.
Both θήσειν τὰ πρυτανεῖά φασί μοι (Ar. Nub.
1180) and δικάσασθαί φασί μοι (id. 1141) are
good Greek : but the latter means, not ‘they
say they will bring an action against me,’
but ‘they speak of bringing an action
against me.’ Cf. Pl. Huthyd. 278 C. ἐφάτην
yap ἐπιδείξασθαι---ἃ τι idiomatic use, which has
come down from forms like Hes. Op. 455,
φησὶ δ᾽ ἀνὴρ φρένας ἀφνειὸς πήξασθαι ἅμαξαν,
‘speaks of making,’ and Hom. 71. 13, 666,
πολλάκι γάρ ot ἔειπε--- φθίσθαι, ‘announced
him a perishing.’ These and similar con-
structions after νομίζω, ἡγεῖσθαι, οἴομαι, φρονῶ,
εὔχομαι etc. are idiomatic survivals from the
fuller predication once conveyed by these
verbs, ‘I am minded for,’ ete., to which the
original use of the inf. as a dative naturally
accommodated itself. The instances of this
use are therefore most frequent in the
earlier stage of the language: οἵ. 71. 2, 401,
εὐχόμενος θάνατόν τε φυγεῖν, ‘for an escape,’
18, 497, ὃ μὲν εὔχετο πάντ᾽ ἀποδοῦναι (‘to pay
all’) ὃ δ᾽ ἀναίνετο μηδὲν ἑλέσθαι (‘refused to
receive anything’). Compare ἐέλπετο κῦδος
ἀρέσθαι (It. 12, 407) with ἐέλπετο νῆας ἐνιπρή-
σειν (15, 701); so Xenophon can say izo-
σχόμενος μὴ παύσασθαι (An. 1, 22) or ὑπέσχετο
μηχανὴν παρέξειν (Cyr. iv. 1, 31). Similarly,
Il. 3, 98, φρονέω δὲ διακρινθήμεναι ἤδη
᾿Αργείους καὶ Τρῶας, ‘I am minded for the
Argives and Trojans being separated.’ Hence
the same construction follows ὄμνυμι, oaths,
which are of the nature of formulae, tending
to preserve archaic constructions, Od. 4,
256, ὦμοσε μὴ μὲν ἀναφῆναι, Hdt. 5, 106,
ἐπόμνυμι μὴ ἐκδύσασθαι κιθῶνα. ‘These and
similar passages seem to have been quite
needlessly altered by recent editors against
the best MSS. e.g. Jl. 20, 84, ἀπειλαὶ &s—
ὑπίσχεο, ᾿Αχιλλῆος ἐναντίβιον πολεμίζειν, Where
the pres. inf. seems sound, being in app. to -
as, ‘the threats you undertook, namely, the
fighting, 21,477 εὐχομένου---πολεμίζειν. Cf.
22, 119: γερούσιον ὅρκον ἕλωμαι μή τι κατα-
κρύψειν, αλλ᾽ ἄνδιχα πάντα δάσασθαι,
where Mr. Leaf says the aor. ‘cannot be
right,’ although in 235 νοέω τιμήσασθαι
he rightly adheres to the MSS. reading.
Several instances also occur in Attic
writers, where the MSS. authority seems
challenged without reason. Thus in all the
Thucydidean passages cited by Mr. Ruther-
ford (1, 11; 4, 24; 4, 80; 5, 28; 7, 21),
the inf. has no subject, and the aor. or pres.
may therefore stand as well as the fut. And
4, 13, in which he finds especial difficulty,
illustrates Thucydides’ usage particularly
well, ἐλπίσαντες τὸ τεῖχος tos μὲν ἔχειν,
ἀποβάσεως δὲ μάλιστα οὔσης ἑλεῖν μηχαναῖς.
Here, as he says, ‘ ἐλπίσαντες 1 the first case
means believe, but for the second clause
hope’; hence he suggests μάλιστ᾽ av. But
it may be observed that ἑλεῖν, which has no
subject, needs not be oblique, and ἐλπίσας
ἑλεῖν, ‘hoping to capture,’ presents no diffi-
culty at all. (Soin Eur. H. F. 746, πάλιν
ἔμολεν ἃ πάρος οὔποτε---ἤλπισεν παθεῖν, we do
THE CLASSICAL REVIEW.
not think ‘ Euripides wrote ἤλπισ᾽ ἂν παθεῖν.
Thucydides rather affects the double con-
struction illustrated above. Shilleto has
well explained this use at ll. 3, ἐνόμισαν
ἐπιθέμενοι ῥαδίως κρατῆσαι, comparing iv. 127,
νομίσαντες φεύγειν τε αὐτὸν καὶ καταλαβόντες
διαφθείρειν, ‘ thinking he was in flight and to
destroy him.’ Yet recent editors have altered
the reading of passages where just the same
use reappears: vii. 42, 3, 6 δὲ Δημοσθένης...
νομίσας οὐχ οἷόν τε εἶναι διατρίβειν, οὐδὲ
παθεῖν ὅπερ 6 Νικίας ἔπαθεν, ‘thinking
delay out of the question, and not being
minded to fall into the same trap as Nicias
had fallen into.’ This construction must be
distinguished from cases like 11. 54, 2: τὴν
ἐπιχείρησιν ἐφ᾽ ἑαυτῷ τε ἐνόμιζεν εἶν at, ὁπόταν
βουλῆται...καὶ τότε καλλίστην γίγνεσθαι,
‘thinking “the attack rests with me, and
that’s the best time for it,’’’ where we have
the oblique of the ‘vivid’ form, which ap-
pears in iv. 10, 3, τοῦ χωρίου τὸ δυσέμβατον
ἡμέτερον νομίζω (ὃ) μενόντων ἡμῶν ξύμμαχον
γίγνεται: οἵ. Shilleto ad i. 127, 1. Both
usages are united in iv. 24, 4 : Ῥήγιον ἤλπιζον
ῥαδίως χειρώσασθαι καὶ ἤδη σφῶν ἰσχυρὰ τὰ
πράγματα γίγνεσθαι, ‘hoping to capture
Rhegium with ease, and ‘then we are
2292 = > Ν
safe” - iv. 9,2, ἐπισπάσασθαι αὑτοὺς
101
ἡγεῖτο προθυμήσεσθαι... ἐκείνοις τε βιασαμένοις
τὴν ἀπόβασιν ἁλώσιμον τὸ χωρίον γίγνεσ-
θαι, ‘thinking to entice them to be eager
(Goodwin, p. 42) and that, if the enemy
force a landing, the place zs in their power’ ;
iv. 117, 1, νομίσαντες μὲν otk ἂν ἔτι τὸν
Βρασίδαν σφῶν προσαποστῆσαι οὐδὲν...
καὶ ξυμβῆναι τὰ πλείω, ‘thinking Β.
would not...and intending to make peace.’
For a similar idiom in Latin ef. ‘facere dix-
erunt,’ ‘putas posse,’ etc. (Roby 1345).
We venture then, to dissent from Mr.
Monro (#. G. 235), ‘the notion of future
time may be given by an aorist or present
inf.’ as well as from Mr. Rutherford (Babr.
l.c.), that ‘ well- known and_acknow-
ledged errors of copying’ are needed to
account for ‘apparent exceptions to the
legitimate construction,’ which we would
state as follows: ἐλπίζω ete. followed by an
ace, and inf. takes the same construction as
any other ‘verb of reporting’; when
followed by a simple inf., that inf. may
either be oblique, in which case it takes the
tense of the direct form, or ‘prolative,’ 1.6.
a direct object, in which case it may be pres.
or aor., and will not denote time (pres., past
or fut.) at all.
W. T. Lenprum.
ON THE ANCIENT LAW OF SEARCHING FOR STOLEN PROPERTY.
Gatus, in dealing with the actio prohibiti
Jurti (111. 189 et seg.), says that the Twelve
Tables enacted (Tab. VIII. 1. 15) ‘ut qui
quaerere velit, nudus quaerat, linteo cinctus,
lancem habens.’ In discussing this, he says
‘quid sit linteum quaesitum est: sed verius
est, aliqguod consuti genus esse, quo necess-
ariae partes tegerentur.’ He mentions two
explanations of the lanx only to condemn
them. But Festus gives a third without
remark: ‘lance et licio (1.6. linteo licio)
dicebatur apud antiquos, quia qui furtum
ibat quaerere in domo aliena, licio cinctus
intrabat, lancemque ante oculos tenebat
propter matrumfamilias aut virginum prae-
sentiam.’ Those who accept this explanation
of lanx say that it was held before the
eyes that the searcher might not be recog-
nized. Now I believe this law is the outcome
of a primitive custom of searching naked
for stolen property. For the sake of decency,
the barest covering was added, but such that
nothing could possibly be secreted in it.
The theft in these cases was counted as
‘manifest’ (Gelly xi, 18, 9), which shows
that this law is the outcome of a state of
things in which the injured man took the
law into his own hands and struck while his
blood was hot. (See Maine’s Ancient Law,
p- 379). Now whatever the meaning and
use of this lanx, common sense demands
that a man who goes to search for stolen
property should have his hands free, and I
fail to see how a man quits the indecent for
the decent state by holding a perforated
plate before his face. I believe that in
place of lancemque ANTE OCULOS TENE-
BAT, which the stupid epitomizer of Festus
gives, the original ran lancemque A LINTEO
SUSPENDEBAT. Thus Festus took the
covering to consist of a girdle suspending a
lanx. Whether Festus’ explanation rests
on any better authority than those rejected
by Gaius, who can say? But the fact that
he had an explanation which satisfied him
sets up a strong presumption that it was,
not necessarily the true one, but a rational
one. Therefore the folly of it as it now
102
appears in the text is probably due not to
Festus the lexicographer, but to Paulus the
epitomizer.
In the parallel passage in Plat. Laws 954
φωρᾶν ἂν ἐθέλῃ τίς τι παρ᾽ γυμνὸς
ἢ χιτωνίσκον. ἔχων ἄζωστος ... οὕτω φωρᾶν,
C. Hermann is prob: vbly wrong in bracketing
ἢ as spurious. This is from an old law, and
in early language neither γυμνὸς nor nudus
bore the secondary meaning they afterwards
acquired, viz. ‘wearing only the tunica.’
Thus in Homer (Od. 6, 222) a man is
ashamed to appear γυμνὸς, but (Od. 14, 489)
not ashamed to appear οἰοχίτων. The dis-
tinction is sometimes kept up in Attic,
sometimes lost. Thue. 1, 6 in contrasting
ἐγυμνώθησαν with διαζώματα ἔ ἔχοντες keeps it.
So does Demosthenes in 21 ὃ 216 μικροῦ
γυμνὸς ἐν TO χιτωνίσκῳ γενέσθαι, by inserting
μικροῦ. Moreover Plato’s searcher is not to
wear a girdle; but when γυμνὸς and pudus
mean ‘ without the ἱμάτιον,᾽ they still imply
that the girdle is retained, as in Vergil’s
ὁτῳοῦν,
THE CLASSICAL REVIEW.
‘nudus ara,’ where nudus means ‘ wearing
the tunica succincta,’ and in Dem. 54 ὃ 9
ἀπεκομίσθην γυμνὸς, οὗτοι δ᾽ ᾧχοντο θοιμάτιον
λαβόντες. Hence in Plato there is a contrast
between γυμνὸς and ἄζωστος. The Greeks
qualified the original requirement of com-
plete nudity in one way, the Romans in
another. In Aristoph. Clouds 497 (Soer.
ἴθι νυν, κατάθου θοιμάτιον. Streps. ἠδίκηκά τι;
Socr. οὐκ ἀλλὰ γυμνοὺς εἰσιέναι νομίζεται.
Streps. ἀλλ᾽ οὐχὶ φωράσων ἐγώγ᾽ εἰσέρχομαι),
there isa play on the two meanings of γυμνός,
Strepsiades in ἠδίκηκα and φωράσων under-
standing Socrates to mean that he will have
to take off girdle and χιτών as well as ἱμάτιον.
I venture to doubt whether Westermann
and Sandys rightly quote Lysist. 150 ἐν τοῖς
χιτωνίοις γυμναί to illustrate Dem. 54 ὃ 9,
since in the Lysistrata γυμνός, as in the law
before us, has its original sense, the χιτωνία
being transparent.
EE. C. MarcHant.
ON SOME PASSAGES IN THE TEXT OF PLATO'S THEAETETUS.
149 D.—xat μὴν καὶ διδοῦσαί ye αἱ μαῖαι
φαρμάκια καὶ ἐπάδουσαι. δύνανται ἐγείρειν TE τὰς
ὠδῖνας καὶ μαλθακωτέρας, ἂν βούλωνται «ποιεῖν,
καὶ τίκτειν τε δὴ τὰς δυστοκούσας, καὶ εὰν νέον ὃν
δόξῃ ἀμβλίσκειν, ἀμβλίσκουσιν ;
The words νέον ὄν are generally held to be
corrupt : Schanz gives no fewer than four-
teen emendations of different critics, besides
his own. Of these perhaps the best is
Heindort’s δέον. certainly the worst is
Naber’s καὶ vavov ἂν δάξη. I conjecture νηδύν
for νέον ov: cf. Eur. Andr. 355—6 ἡμεῖς yap
εἰ σὴν παῖδα φαρμακεύομεν, καὶ νηδὺν ἐξα μ-
βλοῦμεν κ-.τ.λ.
155 E.—yapw οὖν μοι εἴσει, ἐάν σοι ἀνδρός,
μᾶλλον δὲ ἀνδρῶν ὀνομαστῶν τῆς διανοίας τὴν
ἀλήθειαν ἀποκεκρυμμένην συνεξερευνήσωμαι
τος
αὑτῶν ;
The situation is this. Socrates humorously
pretends that the true meaning of Protagoras’
πάντων μέτρον ἄνθρωπος has been intention-
ally hidden away, being communicated to
his pupils only in camera. If we look closely
at the passage, we shall see that the
metaphor is from the chase: this is clearly
the meaning of συνεξερευνήσωμαι: ct.
Homer Odyssey xix 435—436 οἱ δ᾽ és βῆσσαν
ἵκανον ἐπακτῆρες" πρὸ δ᾽ ap αὐτῶν ἴχνι’ ἐρευ-
νῶντες κ ves ἤισαν. But what of the final
αὐτῶν! Except by Campbell, the word is
regarded as corrupt by nearly every critic,
as a glance at Schanz’s apparatus ecriticus
will shew.
Campbell suggests two explanations:
either to take αὐτῶν as recapitulatory of
ἀνδρῶν ὀνομαστῶν, or to make the word
depend on ἐξ in συνεξερευνήσωμαι. Neither
explanation seems consistent with the luci-
dity of Plato’s style. Madvig and Kreyen-
biihl suggest ἰών and μετιών respectively :
they areright i in so far as they see that a parti-
ciple is required, and Kreyenbiihl has done
well in selecting one with associations of
the chase. Plato perhaps wrote ἀστῶν, from
ἀυτῶ, the words ἐρευνήσωμαι αὐτῶν being from
the end of a hexameter. Fragments of
poetry are frequently imbedded in the
Platonic writings ; some will be pointed out
later in this very dialogue. Translate:
‘will you be grateful to me then, if with a
halloo I help you to ferret from its lair
(ἀποκεκρυμμένην) ete.’ The words accord
well with Socrates’ hilarious mood: and
Plato’s fondness for the ‘ venatory metaphor’
is well known (e.g. Republic iv 432 B—E).
156 D.—séoov μὲν οὖν βραδύ, ἐν τῷ αὐτῷ καὶ
πρὸς τὰ πλησιάζοντα τὴν κίνησιν ἴσχει καὶ οὕτω
δὴ γεννᾷ, τὰ δὲ γεννώμενα οὕτω δὴ θάττω ἐστίν.
φέρεται γὰρ καὶ ἐν φορᾷ αὐτῶν ἡ κίνησις
πέφυκεν.
THE CLASSICAL REVIEW.
The difficulties of this passage are well
known. Plato is describing the πάντα ῥεῖ
theory of perception. αἴσθησις and αἰσθητόν
are declared to be the product of the inter-
action between the ego (operating through the
organ of sense) and the object. Now each of
these, the ego, the object, the αἴσθησις and
the αἰσθητόν, is a κίνησις or mode of change :
the two former being slow, the two latter fast
κινήσεις. It is this which the words quoted
are meant to bring out, and even as they
stand the meaning is clear, but the expres-
sion is faulty, as the editors (including
Campbell) admit. Cornarius supplies after
the second οὕτω δή a long clause in which he
completely mistakes Plato’s meaning: see
Campbell’s note. Read, by transposition
and omitting γὰρ after φέρεται, τὰ δὲ γεννώμενα
φέρεται καὶ ἐν φορᾷ αὐτῶν ἡ κίνησις πέφυκεν
«καὶ οὕτω δὴ θάττω ἐστίν. The whole
sentence is then a perfect balance. With ἐν
τῷ αὐτῷ καὶ πρὸς τὰ πλησιάζοντα τὴν κίνησιν
ἔχει contrasts φέρεται καὶ ἐν φορᾷ αὐτῶν ἡ
κίνησις πέφυκεν, and to ὅσον μὲν οὖν βραδύ,
καὶ οὕτω δὴ θάττω ἐστίν forms the right antithe-
sis. The words φέρεται καὶ ἐν φορᾷ αὐτῶν ἡ κίνη-
σις πέφυκεν perhaps formed in the archetype
a single line which fell out and and was
restored after θάττω ἐστίν as a clause of
reason, with γάρ instead of καί.
162 E—re εἰκότι χρῆσθε, ᾧ «εἰ ἐθέλοι
Θεόδωρος ἢ ἄλλος τις τῶν γεωμετρῶν χρώμενος
γεωμετρεῖν, ἄξιος οὐδ᾽ ἑνὸς μόνου ἂν εἴη.
The words οὐδ᾽ ἑνὸς μόνου have rightly
proved a stumbling-block: Bonitz suggests
οὐδ᾽ ἑνὸς Adyou,Madvig οὐδενὸς μνοῦ, Naber οὐδὲ
Μάνου. Read οὐδ᾽ ἑνὸς νόμου, and understand
the phrase agios οὐδ᾽ ἑνὸς νόμου as a Sicilian
NOTES ON GREEK MSS.
103
proverb, meaning ‘not worth a single groat.’
Plato may well have brought the proverb
with him from Sicily or lower Italy, where
νόμος Was common in this sense: see Head’s
Historia Numorum Ὁ. 55 and compare
Epicharmus Frag. 93 (Ahrens) κάρυξ ἰὼν
εὐθὺς πρία μοι δέκα νόμων μόσχον καλάν.
108 E.—ri δέ; οὐ πολλῶν τοὶ Θεαίτητος
μεγάλους πώγωνας ἐχόντων ἄμεινον ἂν ἐπακολου-
θήσειε λόγῳ διερευνωμένῳ ;
The words μεγάλους πώγωνας ἐχόντων have
manifestly a hexametrical rhythm. Perhaps
there is an allusion to some such line as
ἄξιος ἴδρις ἀνὴρ πολλῶν πώγωνας ἐχόντων.
109 Ο.-- πάντως τὴν περὶ ταῦτα εἱμαρμένην,
ἣν σὺ ἐπικλώσῃς, δεῖ ἀνατλῆναι ἐλεγχόμενον.
The insertion of av after ἥν is needless, for
the words clearly refer to some fragment of
dramatic poetry. The poetic ἀνατλῆναι and
especially ἐπικλώσῃς, together with the semi-
poetic εἱμαρμένη; make this inference obvious,
and I believe we have here an allusion to
some such lines as those of Cleanthes :
+ , 3 κ" a Ν , 5 ε ,
ἄγου δέ μ᾽ ὦ Ζεῦ καὶ σύ y ἡ πεπρωμένη
ΕἸ Lal
ὅποι ποθ᾽ ὑμῖν εἰμι διατεταγμένος,
ε ων ΄ > nv Ν Ν /
ὡς ἕψομαί γ᾽ aoxvos: ἢν δὲ μὴ θέλω
κακὸς γενόμενος, οὐδὲν ἧττον ἕψομαι.
(apud Epictet. Man. 53.)
Addressing himself to destiny, some
character in a Greek play (most probably by
Euripides) may have said :
2. ὦ 3,
ἀλλ᾽ ὅπῃ θέλεις ἀγε:
εἱμαρμένην γὰρ ἣν ἐπικλώσῃ τύχη
πάντως ἀνατλῆναι χρεών.
J. ADAM.
IN ITALIAN LIBRARIES.
(Continued from Vol. 111]. page 352.)
THE following article contains notes taken
down in Rome and elsewhere in the earlier
half of the year 1889, during a journey in
Italy under the Craven Trust. Their com-
parative scantiness may perhaps be excused
by the circumstance that the writer had
begun to devote his time to a definite philo-
logical object, which allowed him only occa-
sionally to note down points of general
interest. It has seemed worth while to
print them, without waiting for a possible
opportunity of completion, partly because so
many Italian libraries have so little chance
of possessing a printed catalogue that any
information concerning them must for some
time to come possess value, and also to spare
future travellers the trouble of investiga-
ting out-of-the-way collections that have
been more or less completely looked through
by the writer.
The writer desires to take this occasion of
expressing his obligations to the authorities
of the Public Libraries of Pistoia, Lucca
and Siena. It may be well also to say
definitely, what few people now would be
inclined to believe, that there are no Greek
MSS. in the University Library at Pisa.
104
ΒΙΒΙΙΟΤΒΟΑ VALLICELLIANA.
An account of the steps by which the
library of 8. Maria in Vallicella acquired its
books may be read in Blume, iii. p. 161 sq.
The Greek MSS., taken by themselves, do
not afford much indication of ownership ;
occasional references occur to the Portu-
guese Achilles Statius, the so-called founder
of the library, and to other XVIth cen-
tury personages, such as Fulvio Orsini and
Cardinal Sirleto. These I have gathered
together, wherever they were met. The
collection is not of very great importance,
although the notes that I give here do not
fairly represent it. I wish to reiterate this,
lest any one should be deterred from examin-
ing the library by the thought that its con-
tents had been exhausted. There is a full
but somewhat inconvenient MS. catalogue of
Greek MSS. Facsimiles of seven MSS. may
be found in Blanchinus, Lvangelium Quad-
ruplex, Romae 1748.
], A. 25. A Latin Ms. oF Curysostom adversus
Vituperatores vitae monasticac, to which is prefixed
one quinion, being 10 pages, of a xvth cent. vellum
MS. of EuRIPIDEs, containing Bios εὐριπίδου, hypothe-
sis and Hecusa 1—327. It is in the hand of John
Rhosus.
2. B. 86. 5. xii. On f. 1517. in a monocondylion,
ἰωακεὶμ tepduovaxos, 228v. γεώργιος ἔγραψε ὁ τοῦ
ἡλίου.
3. Β. 70. bomb. ff. 91s. xiv. ArscHYLUS, PRoM.
inc. v. 187, S.c. TH., PERSAE.
4, B. 115. bomb. 5. xiv. GrorGIus PIsIDIUS DE
IMPERIO CONSTANTINOPOLITANORUM, etc. At beg.
Fulvius Ursinus dono dedit | Achilli Statio Lusitano.
5, C. 8. chart. 5. xv. OpuscuLA CAESARII; at the
end, ἐν μουνεμβασίῃ ἔγέγραπτο. On the last page,
abhine DC. annos fuisse putat Card. Sirletus et valde
probat.
6, C. 28. chart. 5. xvi. contains lists of MSS. be-
longing to Sirleto and to the Eeclesia Trevirensis.
7, C. 61. s. xv. Gospets. At end, τὸ τέλος ἧκε
THS παρούσης“ πυξίδος ἄγγελος θύτης διεγράψατο φίλοι.
8. C. 29. chart. s,xvi. ΡΉΟΤΙΙ, IULIANI EPISTOLAE.
f. 54, at end of Photius, ex codice no. xxix. 38 Biblio-
thecae Sfortanae, Romae in Die 8 Februarii MDC.
anno Jubilaci; at end of Julian, omnia quae in hoc q”
sunt in exemplart Ant. Augustini desunt, ego tran-
scripsi ex codice vaticano die 26 Februarvi 1588.
9, D. 43. membr. 11 x 82, ff. 85 5. xi. according to
the Index. 5. Grecortr Magni DraAnocr (Latin).
At the end, 2 ff. in Greek, concerning which a note
at the beginning says Jn fine hujus Voluminis swnt
duae Paginae perantiqui Codicis Graeci quae Indicem
incerti operis continent. They are in the hand of
Paul, Abbott of Grotta Ferrata, the writer of the MS.
Grott. Ferr. B. ἃ. 1. of the year 986. See ‘Notes on
Abbreviations in Greek Manuscripts,’ Oxford 1889,
pp. 33, 94,
10. ΕΞ 29. membr. s. x. Dion. AREOPAG.,
Lea scholia, in a fine minuscule hand above the
ine.
11. F. 10. membr. 93 x 63 in., ff. 252, 27 1]. qua-
ternions 5. x. contains CANONS, S. BASIL AND OTHERS’
EPISTLES ON THE CANONS. Handsome minuscule
above the line, small semiuncial scholia. A leaf of
THE CLASSICAL REVIEW.
contemporary minuscule from another book is in-
serted at the beginning.
12. F. 16. chart. s. xv. contains ff. 2-44 VAR. ECCL.,
45-52 ILIAD Καὶ 332-3 467, 53-60 AnrisropH. ACH.
661-893, 61 sg. Hestop O. 1). THroc. Scur. HER.
with Scholia.
13, F. 17. membr. EVANGELIA, al. a. 1330, very
carefully written in the xiith century style. At
end,
μιχαῆλ ἱερεὺσ ὃ καλόθετος" μῆ ἰουλ. 6 ivd. vy
ἔτους ςὡλ [1330]
14, F. 25. membr. 5. xvi. contains ff. 64-71,
ARIsTOPH. Plutus 1-55.
15. F. 59. 2s. xv. Curysosrom. On the flyleaf,
in a different hand to that of the text, some verses
eis τὴν ἁγίαν τῆς θεοτόκου Kolunaw, after which,
Habui a Card. Sirletto repertum in libro ut aiebat |
scripto abhine ceccccece. annos litteris grandibus |
addendum ad ἐκλογάδιον in hoc libro. There is
a reference to Sirleto also in F. 59. 1.
16, F. 47. membr. 74x 4% in., ff. 547, 29 IL,
quaternions, 5. x. contains CANoNs, written in a
beautiful small minuscule.
17. 68. chart xv.—xvi. has at the end two lines of
cryptography.
18. F. 83. chart. 5. xvi. GkoRGII SCHOLARII
yo
GRAMM. ff. 50 on f. 50 iw στρατηγὸς ἔγραψεν τὸ
BiBrtoy | 6 κυθηριώτη».
ARCHIVIO DI SAN PIETRO.
By the permission of the Chapter of S.
Peter’s, and with the kind assistance of Don
Pietro Wenzel and Don Giovanni Silvestri
I spent a morning among the MSS. that are
preserved in the Chapter Library of S.
Peter’s. I subjoin notes of two manuscripts
that I noticed more particularly.
1, C. 152. chart. xv. 84x 6 in.
ff. 1—94 ArisropH. PLuT. from v. 206, CLouDs,
with scholia and interlin. glosses. Hrs. THEOG. ὁ.
schol., followed by the HomILIEs oF SCHOLARIUS
GENNADIUvs, and other ecclesiastical matter.
At the end:
οἱ παρόντες λόγοι καὶ κανόνες ἠἡγοράσθησαν
διὰ τὸν ἅγιον τάφον, παρὰ τοῦ ὁσιωτάτ' ἐν ἱερο
μονάχοις κυροῦ σιλβέστρου καὶ πρωτοσυγγέλου τῶν
ἱεροσολύμων ἐν ἔτει (ρκη [1620]
7 AN tk
μ ἰου ιζ
ἐγραφ. παρ᾽ ἐμοῦ τοῦ ἁμαρτωλοῦ καὶ ἐλαχίστου
Νον
τῶν ἱερέων γγι[1]καὶ οἰκονόμου τῆς ἁγιωτάτης ap Em
σκοπῇ5 πωγιιαννῆ".
other similar inscriptions in different parts of the
book.
9. H. 45. membr. 9x7 im., ff. 308, s. xiii.
palimpsest, very damaged, contains GALEN, part
of his works.
At the end, partly in tachygraphical writing, τοῦ
φιλοσόφου κυροῦ φιλαγάθου.
PISTOIA.
R. Licto FortTiguERRI.
1. A. ὅδ. membr. 103 x 63, ff. 225, 35 11, s. xv.
contains the Intap; without scholia. Initials of
books illuminated, and a picture of Homer at the beg.
In the same hand as the next. Two inscriptions in
THE CLASSICAL REVIEW.
Latin, on f. 1 and at the end, I was unable even with
the Librarian’s help to decipher.
9, A. 24. THEOCRITUS AND HesIop, in the same
hand as A 55; with the subscription Τέρμα θεοκρίτου
ὑπὸ σωζομένου γραφέντος ἀγαθῇ τύχῃ.
The hand is plain and rather poor.
Upon Sozomenus, and the donation of books of
which these two MSS, are part, reference may be
made to Zaccaria, Bibliotheca Pistoriensis 1752 yp.
29 sq. The other Greek MSS. that are mentioned in
the inventory, 7b. p. 44, were not shown tome. The
Liceo possesses a good MS. Catalogue by Vittorio
Capponi, 1873.
Lucca.
R. BrsuiotecaA PUBBLICA.
1, 1387. chart. s. xv. PRoctus on Puat. Tim.
9, 2502. s. xv. PsatTER. At beg. Ou τὸ δῶρον
vedgitov δὲ πόνος“.
9. 1424. chart. 58. χυ. Euriprprs Ηπο. ELectr.
Belonged to Cesare Lucchesini, as did most of these
MSS.
4, 1254. The notebook of some scholar.
5, 1427.. chart. s. xv. THEocR. IDYLL., PHAL.
Epp., Fastes, Hesrop O. D. Au. In various hands.
6, 1426. chart. xv. Azsop. On the vellum cover,
Iste liber est mei Caroli S. Joannis andre | despiglatis
Cwis flor :—
7, 2335. CiceRoNIs SomNiuM Scrip. ET AEsop.
FABUL. GRAECE. a. 1767.
λατινιστὶ μεταφρασθεὶς | καὶ γραφεῖς παρὰ | φραγκίσκου
ξανερίου | Βογγὶ τῆς | ξυναγωγῆς τῆς Μητρὸς | τοῦ
θεοῦ | ἐν Νεαπόλει] ἀψέζ
8. 13875. A comMEDY, ITALIAN AND GREEK, by
Agostino Ricchi, s. xvii.
SIENA.
R. BrisiiotEeca PUBBLICA.
1. THE GosPELS IN GREEK. Membr. 144 x12, 23
ll., signed quaternions, 5. xi.-xiil. ; illuminated and
splendidly bound in boards covered with silver-gilt
105
plates with pictures in enamel. Said to have come
from the imperial chapel at Constantinople.
9. I. ix. 8. chart. 5. xvi. contains AxscH. S. c. TH.
PrersaE, Hom. BarracH., Musarus, Her. ET
LEAND. (part).
Ὁ. I. ix. 4. membr. 5. xv. Heston, O. ἢ). THEoOG.
Scur. Her., THEocr. ΤΟΥΤῚ. (part). Unsigned,
but in the hand of John Scutariota.
. H. vi. 9. membr. s. xvi. PLUTARCH ep
φιλαδελφιάς. At beg. Luzio Borghesi.
Two other MSS. are of the xviiith century and
unimportant.
INDEX.
DATES.
1330, Vall. 13.
1588, Vall. 8.
1600, Vall. 8.
SCRIBES.
Angelus θύτης, Vall. 7.
Georgius Eliae, Vall. 2.
John Rhosus, Vall. 1.
John Scutariota, Siena 3.
John Strategus Cytheriota, Vall. 18.
Michael καλόθετος Vall. 13.
Neophytus, Lucca 2.
{Paul, of Grotta Ferrata, Vall. 9.]
Sozomenus, Pistoia 1, 2.
OWNERS ete.
Agostino, Antonio, Vall. 8.
Bonghi, Francesco Xaverio, Lucca 7.
Borghesi, Luzio, Siena 4.
Joachim, hieromon, Vall. 2.
Lucchesini, Cesare, Lucca 3 ete.
Monembasia, Vall. 5.
Orsini, Fulvio, Vall. 4.
Philagathus, ὃ. Peter’s 2.
Tlwyuavyy (archbishopric), S. Peter’s 1.
Ricchi, Agostino, Lucca 7.
Sforza (library), Vall. 8.
Silvestro (πρωτοσύγκελλος of Jerusalem), S. Peter’s 1.
Sirleto, Vall. 5, 6, 15.
Spigliati, Carlo san Giovanni Andrea, dei, Lucca 6.
Trevirensis, ecclesia, Vall. 6.
ΠΣ W. ALLEN.
TUCKER’ SUPPLICES OF AESCHYLUS.
The Supplices of Aeschylus, a revised text
with Introduction, Critical Notes, Com-
mentary and Translation, by T. G.
Tucker, M.A., Professor of Classical Phil-
ology in the University of Melbourne, late
Fellow of St. John’s College, Cambridge.
London, Macmillan and Co. 1889. 8vo.
pp. Xxxvil, 228. 10s. 6d.
Tuts edition gives proof of many virtues:
common sense, alert perception, lucidity of
thought, impatience of absurdity, a rational
distrust of MS. tradition, anda masculine taste
in things poetical. The learner who attacks
the play with this commentary will find un-
failing help by the way and acquire much
information before his journey’s end. ‘The
old miserable experiences of the classical
student who wants to understand what he
reads, his lonely fights with difficulties
whose presence the editor has never ap-
prehended, his fruitless quest of a meaning
in notes where the editor has rendered Greek
nonsense into English nonsense and gone on
his way rejoicing, are not repeated here.
Here on the contrary is a commentator who
shares the reader’s difficulties, rescues him
from some of them, warns him of some ex-
isting unperceived, and to tell the truth
invents a good many where none exist.
It is Prof. Tucker’s main concern, as it
must be for an editor of this play, to find
100
out what. Aeschylus wrote; and his ad-
ministration of this province will decide the
value of the book as an original contribution
to learning. He has introduced into the
text, I reckon roughly, about 200 conjectures
of his own. It is thecritic’s chief duty, and
should be his chief pleasure, to commend what
is good ; so I begin with four emendations
which I should call quite certain.
115 (as if our studies were not yet enough
perplexed with conflicting numerations Mr.
Tucker has invented a new one: I ignore
this and cite according to Dindorf). τοιαῦτα
πάθεα μέλεα Opeopeva δ᾽ ἐγὼ λιγέα βαρέα daxpv-
οπετῆ, ἴὴ ip, ἰαλέμοισιν ἐμπρεπῆ, ζῶσα
γόοις με τιμῶ. Whether this means conspi-
cuous for or among dirges, neither suits, and
Dindorf is obliged to translate decens, aptus,
1.6. to render ἐμπρεπής as if it were ξυμπρε-
ays. Mr. Tucker writes ἐμ φερῆ which 1
find most convincing: the sense is just what
one looks for and the error of a common
type.
121. πολλάκι δ᾽ ἐμπίτνω ξὺν λακίδι AL VOl-
σιν HL Σιδονίᾳ καλύπτρᾳ : in the repetition
αὖ 138 αἴνοισιν ἢ. At first this looks like
λίνοισιν ἢ, but since ἢ is not at all appropriate
they conjecture ἠδὲ or λίνοισι καὶ: Mr.
Tucker however comparing Cho. 27 λινό-
φθοροι Aakides proposes ξὺν λακίδι λινοσινεῖ,
an admirable correction.
341—4.,
B. βαρέα σύ γ᾽ εἶπας, πόλεμον ἄρασθαι νέον.
X. ἀλλ᾽ ἡ δίκη γε ξυμμάχων ὑπερστατεῖ.
B. εἴπερ γ᾽ ax’ ἀρχῆς πραγμάτων κοινωνὸς ἣν.
X. aidod σὺ πρύμναν πόλεος ὧδ᾽ ἐστεμμένην.
‘Yes, if she was concerned in the affair at
first’ is nothing to the point: that Justice
was concerned one way or the other there
can be no manner of doubt. Or if there is
an insinuation that the Danaids were first in
the wrong, they cannot afford to ignore it in
their reply. Mr. Tucker alters ἣν to 7,
‘that would have force had J been originally
concerned, but now it is no business of mine,’
an answer which compels the suppliants to
shift their ground ; and we are really nearer
the MS. than before, for in the next verse it
gives ἐστεμμένη : the ν was transposed.
φύλαξαι μὴ θράσος τέκῃ φόβον.
Ν \ / m” ri ed 4 “2
καὶ δὴ φίλον τις ἔκταν᾽ ἀγνοίας ὕπο.
No recorded use of καὶ δὴ is here in place :
Mr. Tucker writes 757: one abbreviation of
καὶ is much confused with ἡ, and ἤδη is so
appropriate to the sententious aorist that
the correction, once made, is obvious.
The following conjectures I select as
THE CLASSICAL REVIEW.
favourable specimens. 154. εἰ δὲ μὴ, μελανθὲς
ἡδιόκτυπον γένος τὸν γάιον...Ζῆνα.. «ἱξόμεσθα :
Wellauer’s ἡλιόκτυπον is generally read, but
Mr. Tucker objects that κτύπος means noise
not merely stroke and that ὀμβρόκτυπος
νιφόκτυπος χιονόκτυπος do not warrant ἡλιό-
κτυπος = ἡλιόβλητος : he writes with great in-
genuity μέλαθρ᾽ és ἡλίῳ στυγούμεν᾽ ὡς τὸν γάιον
k.7.A.: the sense is excellent and the changes
though numerous are all easy : it must how-
ever be remarked that the preposition os is
not found in Aeschylus. 198. τὸ μὴ μάταιον
δ᾽ ἐκ μετώπων σωφρονῶν | ἴτω προσώπων : Mr.
Tucker writes κατωποσωφρόνων which he
compares with such compounds as ἁγνόρυτος
and ἀκριτόφυρτος : certainly former conjec-
tures have small likelihood, Porson’s least of
all. 220 sq. A. Ἑρμῆς ὅδ᾽ ἄλλος τοῖσιν
Ἑλλήνων νόμοις. | X. ἐλευθέροις νῦν ἐσθλὰ
κηρυκευέτω: here ἐλευθέροις seems quite
irrelevant, and Mr. Tucker proposes ἀλλ᾽
εὑρεθεὶς with allusion to Hermes as the god
of εὑρήματα : this to be sure gives the verse
a point, though hardly perhaps of the sort
one looks for after the six foregoing lines.
380. φόβος μ’ ἔχει φρένας δρᾶσαί τε μὴ
δρᾶσαί τε καὶ τύχην ἑλεῖν: Mr. Tucker's
difficulties about ἑλεῖν I share to the full ;
but when he writes τύχην ἐᾶν, to be con-
strued closely with p27 δρᾶσαι, the result
is not a well-balanced phrase. 405. τί τῶνδ᾽
ἐξ ἴσου ῥεπομένων μεταλγεῖς τὸ δίκαιον ἔρξαι ;
a man cannot μεταλγεῖν what he has ποῦ
yet done, so Tournier proposes μ᾽ ἔτ᾽ ἀργεῖς :
Mr. Tucker’s pe ταρβεῖς seems better. 480
sqq. σὺ μὲν, πάτερ γεραιὲ τῶνδε παρθένων, |
κλάδους Te τούτους ail’ ἐν ἀγκάλαις λαβὼν |
βωμοὺς ἐπ’ ἄλλους δαιμόνων ἐγχωρίων | θές:
an anacoluthon mended in various ways, as
by altering σὺ to cod (-- ὅρμα) or τε to γε
or aiy’ to aip’ or λαβὼν βωμοὺς to λαβὲ
βωμούς 7’. But Mr. Tucker further points
out what seems to have escaped notice,
that τούτους cannot well be right when the
Danaids are found still in possession of their
κλάδοι at 506, and he removes two difficulties
by the easy change of τε τούτους to τοιούτους.
632. θεοὶ διογενεῖς, κλύοιτ᾽ εὐκταῖα γένει χεού-
cas: γένει. seems quite harmless, but in M it
is corrected into γένη whence Mr. Tucker
conjectures τέλη, offerings: the word is
appropriate and the error common. 854:
certain restoration there can be none in
this wilderness of ruin, but δύσφρον᾽ ἀνάγκαν
for δύσφορα ναὶ κἀν ought to be right. 907.
διωλόμεσθα ἐπτάναξ πάσχομεν : it is usual to
read ἄελπτ᾽, ἄναξ, πάσχομεν with Robortellus,
but Mr. Tucker’s acer? seems more suitable.
924, ἄγοιμ᾽ ἂν, εἴ τις τάσδε μὴ ᾿ξαιρήσεται:
this weak verse Mr. Tucker alters to μάθοιμ᾽
THE CLASSICAL REVIEW.
ἂν εἴ τις τάσδε μ᾽ ἐξαιρήσεται comparing Eur.
And. 715 ὡς ἂν ἐκμάθω | εἴ τίς με λύειν τῆσδε
κωλύσει χέρας : the change cannot be called
certain, but to me it is very attractive.
1018. ἴτε μὰν dorvdvaxtas μάκαρας θεοὺς
γανάεντες πολιούχους τε καὶ οἱ xedp’ ’Epacivou
περιναίουσιν : Mr. Tucker objects to ‘ city-
gods’ comprising both πολιούχους and οἱ χεῦμ᾽
Ἔ.. περιναίουσιν, and writes ἄστυδ᾽, ἄνακτας :
the meaning οἵ ἄστυ however need not be
more restricted here than in ἀστυγειτονουμένας
286: the tragedians, as Strabo vill. p. 356
observes, often make no difference between
burgh and land. 1063. Ζεὺς ἄναξ ἀποστεροίη
γάμον : this sense of the verb is unexampled,
and Mr. Tucker’s ἀποστέγοι μοι is as likely
as Hartung’s ἀποστρέφοι μοι.
Among the residue of the 200 conjectures
there may very well be some which will seem
more probable to other critics than to me,
but there can hardly be many. Mr. Tucker’s
objections to the vulgate are often acute and
true: some instances I have given; others
are the remarks on σκόπον 647 and κλύουσά
γ᾽ ὡς ἂν οὐ φίλη 718, and the geographical
difficulties raised about 254—7. Hisaltera-
tions seldom fail to give a just and straight-
forward meaning ; and the book is almost
wholly free of those incredible emendations
which consult the apices codiewm and consult
nothing else. There is indeed one extra-
ordinary specimen of this class at 146, where
ἔχουσα σέμν᾽ ever ἀσφαλές is altered into
λέχους ἄσεμν᾽ ἐν "Omid. σφάλασα, a reading
dependent on two fables, both invented by
Mr. Tucker,—that Orion assaulted one of
Artemis’ handmaids named Opis, and that
λέχους ἄσεμνα is Greek for unholy lust; but
this stands almost alone. What vitiates
two thirds of Mr. Tucker’s conjectures is
that despite his professions he takes no
due heed to palaeographical probability.
‘In the present work,’ says he in his preface,
‘there have been assumed as axioms......
(iii.) that the reading substituted on conjec-
ture must approve its claims by satisfying
the conditions of palaeography—as a most
natural source of the incorrect reading.’
But this is just what an average conjecture
of Mr. Tucker’s does not. True, it seldom
sets the MS. utterly at naught, and it is
usually fortified by a parade of uncial type,
the decorative effect of which is often
pleasing ; but it diverges too far from the
ductus litterarum to have any convincing
force. Possible no doubt it is; but half-a-
dozen alternatives are equally possible. We
know of course that the scribes did make
mistakes as bad as those which Mr. Tucker
postulates ; but when such mistakes have
107
once been made they can never be corrected.
Jt may be that εἴρηται λόγος was the source
of εἰθείη Διός, Γλκμων of Στρυμών, δηχθῆναι
πόθῳ of μιχθῆναι βροτῷ, φοβουμένους of φόβῳ
φρενός, διόρνυται ἐς οὗ διορνυμένα, τίς ποτ᾽ οὐ of
τις βροτῶν, στυφελώδεις οὗ φυγάδες, κηπωρικὴν
λαβοῦσ᾽ ἀνεωσμένην θύραν of καλωρα κωλύουσαν
θωσμένειν ἐρῶ ; and if the ghost of ever-living
Aeschylus has uprisen before Mr. Tucker
from Acherusian quarters and has begun to
shed salt tears and to unfold in words that
it was so, well. But if not, then such in-
ventions may indeed convince their inventor,
but to the cold world they are offered in
vain.
When Mr. Tucker’s conjectures are not
palaeographically improbable they are apt to
be causeless and even detrimental. Among
the axioms assumed in the preface are
the following: ‘the reading in the text
must hold its place until such cause to
the contrary can be shewn as will satisfy
a rigidly impartial tribunal. The onus
probandi lies entirely with the impugner
of the text.’ ‘The conditions of dispos-
session are these. It must either be proved
that the reading is an impossibility, or
else that in point of grammar it is so ab-
normal, or in point of relevance so mani-
festly inappropriate, as to produce a thorough
conviction that the MS. is in error. I for
my part should call this much too strict ;
but these are Mr. Tucker’s principles. His
practice is something quite different: in
practice no word, however good, is safe if Mr.
Tucker can think of a similar word which is
not much worse. 180 sq. ὁρῶ κόνιν, ἄναυδον
ἄγγελον στρατοῦ. | σύριγγες οὐ σιγῶσιν ἀξονήλα-
τοι: Mr. Tucker removes the stop after
στρατοῦ and alters οὐ to οὗ. ‘The ΜΗ.
reading ov,’ says he, ‘causes an asyndeton
which can only be explained by a protracted
pause after στρατοῦ, while Danaus is awaiting
further developments.’ Five lines later on
we learn that ‘Danaus must have paused
several times in this speech, commenting
from time to time on the further progress of
the Argives,’ so there is an end of that
objection: proceed to the next. There is
none. There are assertions, possibly true
but on the editor’s principles irrelevant, that
the new reading is better; and that is
all. 418 sqq. φρόντισον καὶ γενοῦ πανδίκως
εὐσεβὴς πρόξενος: τὰν φυγάδα μὴ προδῷς κ-τ.λ.:
Mr. Tucker removes the colon and writes
προδούς. ‘The change is a mere trifle, but it
is a trifle for the worse, since the μηδ᾽ ἴδῃς
which follows tallies rather better with μὴ zpo-
δῷς than with γενοῦ πρόξενος μὴ προδούς : and
when Mr. Tucker says that προδούς is given
108
on the same obvious ground as that of the
change ἐκδῷς to éxdovs in 340, he quite mis-
apprehends that obvious ground, which is, if
I must explain it, that the question πῶς is
more properly answered by a participle than
by a finite verb. 517 sqq. ἐγὼ δὲ λαοὺς Evy-
καλῶν ἐγχωρίους | στείχω, τὸ κοινὸν ὡς ἂν εὐμε-
νὲς τιθῶ. [ καὶ σὸν διδάξω πατέρα ποῖα χρὴ
λέγειν : Mr. Tucker puts a comma after τιθώ
and writes διδάξων. Now στείχειν is a neces-
sary preliminary to ξυγκαλεῖν because the
λαοί are away in the city, whence ξυγκαλών
στείχω: to διδάσκειν it is not a necessary
preliminary, for Danaus stands by the
speaker’s side: better then διδάξω than
στείχω διδάξων. If we had διδάξων I would
not take the trouble to alter it, because
the superiority of διδάξω is not worth the
change ; but superior it is. 606. ὥστ᾽ ἀνη-
βῆσαί pe γηραιᾷ φρενί: Mr. Tucker writes
γηραιὰν φρένα because the dative would imply
that it is the aged heart which makes the
speaker grow young again, and because if we
are to express the part or respect in which
the rejuvenation takes place we must use the
accusative: then γήθησε δὲ θυμῷ, for instance,
means ‘he rejoiced because of his soul.’
935. τὸ νεῖκος δ᾽ οὐκ ἐν ἀργύρου λαβῇ | ἔλυσεν:
it occurs to Mr. Tucker that λαβῇ and
βλαβῇ look much alike and both make sense ;
the MS. gives one, therefore Aeschylus wrote
the other. But this method cannot well be
avowed, so he looks round for a stone to
throw at λαβῇ, and the first that comes to
hand is this: “AaBy is avery common word
and always means either a handle ora grasp.
We have no authority for treating it as=
λῆψις. Very good; but at 674 we had these
excellent remarks on φόρος (which is a very
common word and always means tribute, and
which we have no authority for treating as
= φορά) : ‘just as τόκος and πρόσοδος had a
general meaning before and besides interest
and revenue, so φόρος had a general meaning
before and besides ¢tribute. The special
meaning is the only one in prose, but not in
verse. dopa is both tax and crop, and φόρος
should be given the same values. Indeed
weareapt to insufliciently remember thesound
(and etymological suggestion) of Greek words
to Greek ears. dopovs=bearings and yas is
sufficient definition.” Why then are we not
to say that AaBy=taking and ἀργύρου is
sufficient definition? because φόρους has the
luck to be a conjecture and λαβῇ the mis-
fortune to stand in a MS. 961. μονορρύθ-
μους δόμους : ‘ we can hardly speak of a house
havinga ῥυθμός...... a house cannot even meta-
phorically have a ῥυθμός᾽ : 1.6. Mr. Tucker,
preoccupied with the interests of his conjec-
THE CLASSICAL REVIEW.
ture povoppvpous, has omitted to look out
ῥυθμός in the dictionary ; just as at 533 he
asserts, to recommend his proposal there,
that ‘rhythm points to a pause at yévos,’
forgetting that if so, then in the antistrophic
verse 526 rhythm points to a pause in the
middle of the word τελειότατον. But when
one hangs one’s criminals first and tries
them afterwards, a flaw in the indictment
is of no practical consequence.
The emendations of scholars fare no better
than the readings of the MS. if their place
is wanted for a conjecture of the editor's
own. Again and again in passages which
we all thought had been corrected long ago
Mr. Tucker proffers another solution, not
better but newer, and promotes it, with
rigid partiality, to the text. 56. γνώσεται δὲ
λόγους τις ἐν μάκει : Mr. Tucker very properly
objects to μάκει standing alone and alters
λόγους to χρόνου. Can he not see that before
we bestow a thought on this he must de-
molish Martin’s λόγου which removes the
difficulty so much more easily and moreover
is clearly what the scholiast read? Τέ.
δειμαίνουσα dodovs, usually amended to
φίλους: “ but the o preserved in M can scarcely
be accidental’ says Mr. Tucker, and writes
ποιμαίνουσα φόβους : the δ, the ε and the A
can all be accidental, but not the o. 164.
κοννῶ δ᾽ ἄγαν γαμετουρανόνεικον : Mr. Tucker
writes γαμετών τῶν οὐρανοοίκων. It ought of
course to be γαμετᾶν ; but apart from that,
in what respect is this conjecture not inferior
to the yaperas οὐρανόνικον of Victorius? 229.
οὐδὲ μὴ ᾽ν ἽΔιδου θανὼν | φύγῃ μάταιον αἰτίας
πράξας τάδε : this is corrupt, but Mr. Tucker’s
μάταιον αἰτίαν is a conjecture which has no
excuse for existing: it means the same as
Tournier’s ματαίους αἰτίας and comes no nearer
the MS., while Schuetz’s ματαίων αἰτίας is of
course more probable than either. 271 sq.
ἔχον δ᾽ ἂν ἤδη tam’ ἐμοῦ τεκμήρια γένος τ᾽ ἂν
ἐξεύχοιο καὶ λέγοι πρόσως : Robortellus emends
λέγοις πρόσω : for ἔχον δ᾽ ἂν the second hand
in M gives yp. ἔχουσαν : now this second hand
collated M throughout with its original and
corrected it thence, and pure nonsense like
ἔχουσαν cannot be a conjecture: we there-
fore take this for a foothold, write ἔχουσα δ᾽
with Heimsoeth, and all is clear. The
reader now knows how to estimate Mr.
Tucker’s assertion ‘the correction of these
lines must start with ἔχον δ᾽ ἂν. It is con-
trary to all the principles of criticism that
this should be an error for ἔχουσα δ᾽. ἔχων
is the only rational correction.’ Then he
goes on ‘either λέγοι or ἐξεύχοιο 1S wrong,
and the omission of © from the former is less
likely than that of T from the latter.’ But
THE CLASSICAL REVIEW.
no: nothing in the world is more likely than
the omission of o before z. ‘ Next, what
τεκμήρια has the king given? He has made
a statement, but he neither offers nor needs
to offer proofs.’ First Mr. Tucker mistrans-
lates τεκμήρια and then declares it corrupt on
the strength of his mistranslation : it means
here what it means in Ag. 352, testimony.
All this leads up to the reading ἔχων δ᾽ ἂν ἤδη
Tam ἐμοῦ τις ev μέρει | γένος τ᾽ ἂν ἐξεύχοιτο
καὶ λέγοι τορῶς: and to write ἔχουσα δ᾽ for
ἔχον δ᾽ ἂν is contrary to all the principles of
criticism! 502. καὶ ξυμβόλοισιν οὐ πολυστο-
μεῖν χρεών : Mr. Tucker's κἀν ξυμβολαῖσιν is
well enough, only Valckenaer’s ξυμβολοῦσιν
is much better, and Mr. Tucker finds nothing
to say against It. 515. σὺ καὶ λέγων εὔφραινε
καὶ πράσσων φρενί: we all read φρένα with
Heath, against which Mr. Tucker can only
say that εὔφραινε does not ‘ require’ an object:
the rest of the note is a conscience-stricken
apology for the demerits of his own conjec-
ture xept. 744. ἔπλευσαν ὧδ᾽ ἐπεὶ τάχει κότῳ :
for these corrupt words Mr. Tucker writes
ἐπιτυχεῖς σκοποῦ ; but if we enquire why this
is to be preferred before conjectures so much
easier and more attractive as Turnebus’
ἐπιτυχεῖ κότῳ OY Weil’s ἐπικότῳ τάχει, the only
reason that can be rendered is Mr. Coventry
Patmore’s: ‘Say, how has thy Beloved
surpassed So much all others? She was
mine.’
But to shew what Mr. Tucker can do and
dare on behalf of his emendations the follow-
ing example is, as King Pelasgus would say,
οὐχ ὑπερτοξεύσιμον. In 137 8ηῳ. we read
τελευτὰς δ᾽ ἐν χρόνῳ πατὴρ ὃ παντόπτας πρευμ-
ενεῖς κτίσειεν, σπέρμα σεμνᾶς μέγα ματρὸς εὐνὰς
ἀνδρῶν, ἐέ, ἄγαμον ἀδάματον ἐκφυγεῖν. Mr.
Tucker objects to μέγα and alters μέγα
ματρὸς to με δάμαρτος : σεμνᾶς δάμαρτος Means
the august bride of Zeus, the πατὴρ
παντόπτας Who has just been mentioned.
Well, this is very pretty ; but unluckily the
lines σπέρμα σεμνᾶς κτλ. are iterated at 151
sqq., Where they are preceded by the mention
not of Zeus but of Artemis, so that σεμνᾶς
δάμαρτος will there mean the august bride
of Artemis, a discouraging result. Mr.
Tucker’s expedient to avoid this mishap is
of a sort that strikes criticism dumb. He
silently prints Δίας instead of σεμνᾶς : not as
a conjecture, but as if it were the MS
reading.
Here I have given proofs enough of the
disasters which attend us when we desist
from the pursuit of truth to follow after
our own inventions. Thus much it was
necessary to say, because the many students
who will I hope resort to this edition for
109
help and instruction must be warned that
they will find not only what they seek but also
a good deal which they are not to believe.
The book however in spite of its faults is
the most useful edition of the Supplices we
have. The purely explanatory part of the
commentary does not contain very much that
is absolutely new, and this is well; for it is
really a far more venturesome thing, if cri-
tics would but understand it, to propose a new
rendering than a new reading. Among the
most interesting notes are those on 189
ἀγωνίων, 472 ἐκπράξω χρέος, 009 θυμέλαι, 691
πρόνομα, 1071 δίμοιρον. I mention one or
two miscellaneous points which are wrong
or doubtful, not that I think them impor-
tant but because the editor may like to cor-
rect or reconsider them. 19: Mr. Tucker
defends γένος τετέλεσται by Pind. Pyth. ΤΥ
256, τόθι yap γένος Εὐφάμου φυτευθὲν λοιπὸν
ἀεὶ τέλλετο : τέλλετο and τετέλεσται however
do not come from the same verb. 167:
Mr. Tucker proposes a difficult interpretation
because he does not observe that δικαίοις
means fitting. 400: it seems rash to con-
jecture κἀλλοῖον when this adjective, so com-
mon in prose, is found in tragedy at one
place only and makes nonsense there. 503:
there can be no cause for Mr. Tucker’s ren-
dering of ἄγοντας unless it be that a con-
torted interpretation is better than a plain
one. 534: by what artifice can νέωσον
evppov’ αἶνον be made to mean νέωσον αἶνον
ὥστε εὔφρων εἶναι ἡμῖν ἢ 556: Mr. Tucker
supports his conjecture ἐγκυκλουμένα by Ovid
met. 1 730 ‘profugam per totum circuit
orbem’: where did he find this reading and
how does he construe it? 604: ‘we might
read ὅτι πληθύεται᾽ : no: a pyrrhic cannot
constitute the fourth foot of an iambic
senarius. 924: the suggestion μοι ᾿ξαρνήσεται
is of course impossible : it is not certain that
even μοὐξαρνήσεται would be used by Aeschy-
lus. 1035: Mr. Tucker is here involved in
some confusion: we have 7p. B coming be-
fore jp. α΄, and in the note we read of ‘the
other ἡμιχόριον ’ though we have hitherto had
no ἡμιχόριον at all.
The translation is written with vigour and
adroitness, and its rhythm is often admir-
able. Here and there, chiefly in the sticho-
mythia, are crude phrases, and one or two so
afflicting as‘ horde of males’ for ἄρσενος
στόλου. παντὶ σθένει in 147 and ἄτερθε
πτερύγων in 782 are left untranslated ;
ἐφορεύοι in 627 is very freely rendered ‘ fur-
ther’; ‘lea αὖ 722 seems meant for ‘lee’ ;
and at 966 occurs the comical misprint, ‘ for
thy good deeds mayst thou have thy fill of
food, Pelasgian lord.’ A. E, Housman.
110
DR.
Dr. Ruruerrory’s contributions to classical
learning have been marked hitherto by
striking vigour and originality. The book
now under review is distinguished no less by
conspicuous power of generalization, concise-
ness and force of style, and acuteness of logic.
But the thesis which he now seeks to estab-
lish is far more important than any which
he has yet handled, and has a far wider
scope than any of his previous works have
aimed at. Dr. Rutherford will have made
an epoch in classical learning if he succeeds
in making good his case. If he is right,
Thucydides wrote, or may have written, in a
style as simple and direct as that of Xeno-
phon ; all the characteristic Thucydideanisms
are foreign accretions, and should be cut
away. As we have Thucydides now ‘there
is hardly a page which does not supply an
instance of a sentence violating every law of
a sentence, but still regarded as justifiable
in Thucydides, who for his great merits of
another kind is to be forgiven occasional
lapses into utopian syntax’ (p. xiv). But
we have not the work of Thucydides as it
was written, or anything like it. According
to the editor nearly all the violations of
normal grammar may be explained on one
principle. In the text of Thucydides have
become embedded thousands of the glosses,
comments, adscripts (to use the general term
which he employs to include them all) of the
scholiasts, copyists, ancient critics, and all
and sundry who have had anything to do
with his great work.
Dr. Rutherford does indeed point out (and
very acutely) certain marked characteristics
which distinguish the style of Thucydides
from that of (say) Xenophon. These are,
broadly : (1) the idiom by which almost any
verb may in the active be paraphrased by
ποιεῖσθαι and some substantive expressing
the action of the verb —an idiom which,
when its boundaries are somewhat enlarged,
has a tendency to elude us, and disguise
itself as grammatical license ; (2) his manage-
ment of participles ; (3) the frequent use he
makes of the indirect reflexive pronoun in
the plural. Moreover he allows that there
is such a thing as style, such a thing as
genius modifying language, and giving it the
1 The Fourth Book of Thucydides, a Revision of the
Text, illustrating the principal Causes of Corruption
in the Manuscripts of this Author, by WM. GUNION
RourHerForp, M.A., LL.D., Head-Master of West-
minster ; author of the New Phrynichus and editor of
Babrius. london, Macmillan, 1889.
THE CLASSICAL REVIEW.
RUTHERFORD’S THUCYDIDES, BOOK IV.!
special impress of a great writer’s mind.
But nine-tenths of the so-called violations of
the common law of grammar, which are
supposed to be characteristic of Thucydides,
arise, according to the editor, from the fact
that into the very warp and woof of his
language, as it has come down to us, has
been wrought the adscript pattern of the
copyists and scholiasts. Were it not for this
his style would be as free from grammatical
license as that of Plato or Demosthenes.
Indeed Dr. Rutherford claims for Thucydides
a singular preciseness in the use of gram-
matical forms, and illustrates his case by
some very acute remarks on passages in the
fourth book. For instance in c. 98 σπένδουσιν
is defended as an instance of extreme logical
exactitude, because ‘ one side can only bid the
other σπένδειν, 1.6. do their part in the
common ceremony, not σπένδεσθαι, ἐ.6. do the
part of both.’ So dvaravovres in ὁ. 11 has a
reason in strict logic, for it is the relieving
party who make the attack. A like pre-
cision is claimed for the active ξυνεπῆγον,
6. 79, and for the reciprocal middle δια-
κινδυνεύεσθαι, c. 19; and ine. 71 Dr. Ruther-
ford finds in the word ἐφεδρευόντων an apt
reference to the ἔφεδρος who sat by in the
gymnic contests, ready to step in and con-
tend with the victor.
How, then, did it come to pass that this -
most precise of writers thus became the
victim of a set of circumstances which have
presented to us as his work something which
certainly cannot be called precise? Every
one admits that many adscripts have forced
their way into the texts of the ancient
Greek writers, and in Thucydides especially
numbers have been pointed out by Dobree,
Krueger, Badham, Herwerden, and above
all by Cobet. The question which at once
presents itself is, how did it happen that the
text of Thucydides suffered so much more
from this source of corruption than that of
Xenophon, Demosthenes, or Plato? I cannot
find any really satisfactory answer to this
question in Dr. Rutherford’s book. When
he writes :
In Thucydides especially this kind of corruption
has escaped notice more easily because of his un-
deserved reputation for obscurity and clumsiness of
expression,
surely we have here something like that
form of petitio principii which logicians call
‘the vicious circle.’ Whence came the un-
deserved reputation? Was it not from the
adscripts? But what peculiar power of
THE CLASSICAL REVIEW.
attracting adscripts had the text of Thucy-
cides ?
In his text Dr. Rutherford prints the
supposed adscripts in uncials in the margin,
and certainly in a great many instances their
expulsion from the text is a great improve-
ment to the style. A good example occurs
in 126 (I inclose the adscripts in square
brackets) : γνώσεσθε τὸ λοιπὸν ὅτι ot τοιοῦτοι
ὄχλοι τοῖς μὲν τὴν πρώτην ἔφοδον δεξαμένοις
ἄπωθεν [ἀπειλαῖς] τὸ ἀνδρεῖον μελλήσει ἐπι-
κομποῦσιν, Where ἀπειλαῖς is evidently a gloss
on μελλήσει used in its thoroughly Thucy-
didean sense of ‘a demonstration’ of feeling,
as opposed to action. Again, in the speech
of Hermocrates, c. 63, he detects a serious
interpolation of nineteen words from κατ᾽
ἀμφότερα to εἰρχθῆναι, and fairly asks com-
mentators ‘can they honestly render 76
ἐλλιπὲς εἴργεται into English or any other
tongue?’ In some cases the interpolation is
harder to detect because it has undergone
some change—generally a change of case—
to enable it to take its place as an integral
part of the sentence. Inc. 34 all the MSS.
give αὐτοὶ τῇ τε ὄψει ἱτοῦ θαρσεῖν] τὸ πλεῖστον
εἰληφότες πολλαπλάσιοι φαινόμενοι. Dobree
emended πλεῖστον to πιστόν. Dr. Rutherford
holds that τὸ θαρσεῖν was a gloss on πιστόν,
which was changed to τοῦ θαρσεῖν to make it
fit into the sentence. In the same chapter
ἀποκεκλημένοι τῇ ὄψει [τοῦ προορᾶν] was
originally ἀποκ. τῆς ὄψεως, but when the
gloss got into the text τῆς ὄψεως was changed
to τῇ ὄψει to procure a construction.
Most interesting are those additamenta
which are betrayed by the presence in them
of some idiom unexampled in classical usage
but prevalent in late Greek. Of these the
most remarkable are ore ἐπεφεύγει, 133, where
the pluperf. is used as a simple past, as in
later Greek; διὰ τὸ περιέχειν αὐτὴν, 102,
where διὰ must bear its post-classical mean-
ing of ‘for the purpose of’; so in 40 δι
ἀχθηδόνα is a gloss because the meaning
required is not the classical i distress, but
the post-classical to cause annoyance ; and in
67 διὰ τῆς τάφρου is foreign matter because
the context requires for διὰ the late meaning
of by or along. With like certainty we may
reject the adscript in 120, περὶ δὲ τὰς ἡμέρας
ταύτας [ais ἐπήρχοντο] because ‘even if
Thucydides may have used ἐπήρχοντο, so far
as form goes, he undoubtedly no more used
it in the late sense of discuss than he used
διὰ in the late sense of to cause and along.’
Dr. Rutherford even gives us marks
whereby we may detect the intruder into
the text. A very common way of intro-
ducing a marginal or interlinear note is by
151}
a relative pronoun, adverb, or conjunction,
either simple or (far more commonly) com-
pounded with περ. Suspect a clause beginning
ὅπερ, ὥσπερ, οἷσπερ, ἵναπερ. Thus in 48
ἵναπερ τὸ πρῶτον ὥρμηντο Should be expelled
from the text, not corrected by a change of
ἵναπερ to ἔνθαπερ or οἵπερ.
Sometimes the adscript gets into the wrong
sentence, or the wrong part of a sentence,
as in 66, ot δὲ φίλοι τῶν ἔξω τὸν θροῦν
αἰσθόμενοι φανερῶς [μᾶλλον ἢ πρότερον] καὶ
αὐτοὶ ἠξίουν κιτ.λ., Where the adscript really
belongs to the following sentence, γνόντες οὐ
δυνατὸν τὸν δῆμον ἐσόμενον.
It is a pity that Dr. Rutherford almost
completely confines himself to the limits of
Bk. IV. in illustrating his theories. One
would have been glad to know whether he
regards as a gloss the οὐ in μᾶλλον ἢ οὐ,
where it is apparently redundant, as in
μᾶλλον ἢ ov τοὺς αἰτίους, IIT. 36. It is very
interesting to read Thucydides afresh and
watch for places where Dr. Rutherford would
resort to excision. One feels at once a
bloodthirsty desire to wield the weapon he
has put into our hands; we cry φονᾷ φονᾷ
νόος ἤδη, and we see everywhere τομῶντα
πήματα to which we will keene no incanta-
tions. To take Pericles’s speech in the second
book, I fancy Dr. Rutherford would cer-
tainly expel from the very beginning of it,
c. 35, ἀγορεύεσθαι αὐτὸν and πιστευθῆναι, and
probably αὐτῶν after ὑπερβάλλοντι. 1 would
infer from the same speech, c. 39, that
Dr. Rutherford goes too far when he says
‘the text of Thucydides is dotted over with
Λακεδαιμόνιοι and ᾿Αθηναῖοι in every case
and every construction, none of which he
ever wrote.’ If the scholiasts and copyists
were so prone to append these names by way
of explanatory comment, why do they not
always append them? In ec. 39, if any-
where, we should have expected οἱ Λακεδαι-
μόνιοι to have crept into the sentence οἱ μὲν
ἐπιπόνῳ ἀσκήσει εὐθὺς νέοι ὄντες τὸ ἀνδρεῖον
μετέρχονται, for surely no one will believe
that any person capable of writing any com-
ment at all could fail to see that the refer-
ence here is to the Lacedaemonians, and yet
the addition of the name would certainly
add to the clearness and antithesis of the
passage.
I fear that if we apply Dr. Rutherford’s
knife impartially in pruning away from the
text what is ungrammatical according to
normal usage, we shall find that many of the
finest things in Thucydides will vanish. The
following is a passage, II. 53, than which
there are few more impressive in Thucydides :
θεῶν δὲ φόβος ἢ ἀνθρώπων νόμος οὐδεὶς ἀπεῖργε;
112
τὸ μὲν κρίνοντες ἐν ὁμοίῳ καὶ σέβειν καὶ μὴ; ἐκ
τοῦ πάντας ὁρᾷν ἐν ἴσῳ ἀπολλυμένους, τῶν δὲ
ἁμαρτημάτων οὐδεὶς ἐλπίζων μέχρι τοῦ δίκην
γενέσθαι βιοὺς ἂν τὴν τιμωρίαν ἀντιδοῦναι, πολὺ
δὲ μείζω τὴν ἤδη κατεψηφισμένην σφῶν ἐπι-
κρεμασθῆναι, ἢν πρὶν ἐμπεσεῖν, εἰκὸς εἶναι τοῦ
βίου τι ἀπολαῦσαι. It will be seen that we
are at once confronted with a glaring viola-
tion of the strict laws of construction—a
most characteristic Thucydideanism. At
first sight the passage seems to illustrate
most forcibly Dr. Rutherford’s theory. The
κρίνοντες Clause and the οὐδεὶς ἐλπίζει clause
are those which interrupt the construction
and mar the grammar. The one would be a
most apt explanatory note on θεῶν φόβος and
the other on ἀνθρώπων νόμος. But where are
we to stop? These supposed adscripts are
inextricably bound up with what follows.
We must refer to the commentator the whole
sentence down to ἀπολαῦσαι, and the result
is that by far the most memorable senti-
ment in the whole chapter belongs, not to
Thucydides, but to the scholiast. I can
think of only three ways in which Dr.
Rutherford could meet the difficulty thus
raised. He may maintain that the theory
of the adscript is here inapplicable ; but it
seems to fit the passage toa nicety. Or he
may show that the words of the comment
may be disentangled from those of Thucydides
without carrying away with them most of
that to which the passage owes its glow and
strength; but in the present case this seems to
me to be impossible. Or finally, he may hold
the theory that among the many scholiasts
there may have been one or two capable of
rising to the height of feeling and expression
required by such a sentence as πολὺ δὲ μείζω
τὴν ἤδη κατεψηφισμένην, κιτιλ. He may hold
that there may have beena scholiast capable
of doing for the text of Thucydides what
Carlyle has done for the letters and speeches
of Cromwell, capable of emphasizing and
enhancing the beauties of the expression, and
deepening and elevating the train of thought.
Then I ask where are the writings of this
KapAvAvos? How did this genius content
himself with the humble labours of a com-
THE CLASSICAL REVIEW.
mentator when he possessed original faculty
of so high an order? I believe there never
was a scholiast capable of conceiving or ex-
pressing this sentiment, though considera-
tions of grammar only would point to it as
an undoubted scholiastic comment.
Dr. Rutherford’s work teems with evi-
dences of his high and deep scholarship, and
with instructive and suggestive comments on
Greek usage. He has embodied in his text
many acute conjectures of recent scholars,!
and several of his own, which, though as a
rule quite convincing, yet deal with such
minute points that it would not be very in-
teresting to record them here. He has un-
doubtedly detected many adscripts which
have eluded the vigilance of even Cobet
himself, and has thus proved that the text
of Thucydides has suffered more than has
been hitherto suspected from this particular
source of corruption. When he writes
(p. XXvil.) :-—
His style is simple but powerful, a fitting weapon
for a vigorous understanding dealing in an unaffected
way with events and the lessons to be derived from
them,
it seems to me that he has not yet succeeded
in showing that we may fairly call the style
of Thucydides simple. We shall be able to
pronounce a more decided opinion when he
deals with the question more at large—as I
hope he will—and takes his illustrations
from the whole work of Thucydides. One
feels however that it would be unfair to ᾿
claim the immediate performance of so large
a task from one who is the Head Master of
a great public school, and who has already
begun to turn the sod of another new wide
and fertile field of philological labour,
R. Y. TYRRELL.
1 There are two conjectures by the Editor of this
Review published in the Journal of Philology, vol. vii.
p. 234, which might well have found a place among
these. For προσέλαντες in iv. 7 Professor Mayor
would read προσειλήσαντες, ‘having penned up,’
which he defends by comparing Hom. J. x. 347,
Eur. Hel. 445; and for σκαιότητα (al. νεότητα), ib.
80, he conjectures σκληρότητα, a word which ex-
presses ‘the unconquerable determination and stiff-
neckedness of the Helots.’
COUAT’S ARISTOPHANES AND THE ANCIENT ATTIC COMEDY.
Aristophane et ancienne Comédie Attique, par
A. Couat, Recteur de |’ Académie de Lille:
Paris, 1889.
THosE who have read M. Couat’s tude
sur Catulle and his Poésie alexandrine sous
les trois premiers Ptolémées will expect much
pleasure and instruction froma work by the
same hand on the great Athenian wit. And
they will not be disappointed. A French
scholar has perhaps peculiar advantages in
dealing with the genius who from one point
THE CLASSICAL REVIEW.
of view suggests Rabelais and from another
Moliére. But M. Couat does not seek to
gain for his work an adventitious attractive-
ness by reading the past into the present, or
the present into the past. ‘Athénes,’ he
writes, ‘n’est pas la France, et en cherchant
parmi nos écrivains, je n’y vois pas un
Aristophane.’ In the introduction he traces
the rise and growth of comedy. In the
subsequent two books consisting of seven
chapters he deals with the materials which the
comic poets found ready to their hands. This
comprises of course gurequid agunt homines ;
and hence the work is nothing less than a
study of Athenian life during the period of
the Old Comedy from B.c. 446 to 392—
its institutions, public men, social questions,
religion, education, and manners. In learn-
ing the writer is quite abreast of modern
scholarship, and he has his full share of that
charm of style which seems to be the birth-
right of every Frenchman whatever be the
subject of his essay. Especially brilliant is
his description of the fierce struggle for
literary existence at Athens in the time of
Aristophanes. The poets were innumerable.
The chances of getting the public ear were
few. It was a triumph even to obtain a
hearing for one’s work. To be known at all
a poet must be among the first poets of his
time—entre la lumiére éclatante du thédtre de
Dionysus et Vobscurité profonde il n’y avait
pas dintermédiaire. The relations between
Aristophanes and Euripides are touched inci-
dentally with much originality and clearness.
M. Couat regards the Aristophanic attack
on Kuripides as but an episode in the war
against Socrates and the Sophists; an
ancient comic poet called the tragedies of
Euripides σωκρατογόμφους. Of the stage of
Euripides, as satirised by Aristophanes,
with its lame and ragged kings, poor but
fluent, M. Couat writes, c’est la cour des
Miracles des héros. Before Euripides
Tragedy had to do only with the grand
adventures and the dignified sufferings of
heroes and heroines; he enlarged it to
embrace the troubles of every-day life and
the loves and jealousies of ordinary women.
The women of Sophocles suffer, those of
Kuripides souffrent. ΝΜ. Couat sums up the
influence of Kuripides on Tragedy in a
delightfully French expression, avec Luripide
les événements tragiques ne se passent plus dans
un palais, mais dans un alcéve. But he is
far from adopting the views of Aristophanes
1 Some verses preserved from the Melanippe appear
to contain the retort of Euripides on his persistent
assailant (frag. 495, Nauck).
NO. XXXI. VOL. IV.
113
on this subject. It must be remembered
that in the time of Aeschylus, and for the
first fifty years of the life of Sophocles, prose
was unknown at Athens. ‘The first play of
Euripides of which we know the date was
produced when the poet was forty-two years
of age, and when prose had come to Athens
with the Sophists, and the reign of King
Logos was beginning.” With them came ina
tendency towards introspection and analysis,
and Euripides saw how powerfully this grow-
ing taste might be appealed to by remodelling
the method of Tragedy. Hitherto there had
been no such thing as the analysis of human
passion in poetry. Homer had told us how
the love of Helen for Paris had been the
cause of the Trojan war. We had read how
the passion of Haemon for Antigone had
precipitated the doom of the House of Lab-
dacus, and how the lewdness of Clytaem-
nestra brought ruin on the Atreids. We
had seen the work of Aphrodite at her
terrible play,? but neither Homer nor
Aeschylus nor Sophocles analysed the passion,
or observed its symptoms. Euripides was the
first who saw that the passion of love was
worth watching for itself. Noone will now
hesitate to admit that the bounds of Tragedy
were widened when Medea and Phaedra
walked the stage, or will sympathise with
the boast which Aristophanes puts into the
mouth of Aeschylus—that no one can say he
ever brought a love-sick woman into his
plays.*
The whole debate between Aeschylus and
Euripides in the /rogs has now for us ἃ new
and special interest, because it resolves
itself to a great extent into a question which
is now agitating our critics—how far realism
is reconcilable with art. When Aeschylus
condemns the portrayal of the foul passions
of Phaedra and Sthenoboea, Euripides
triumphantly asks, ‘ But do not the passions
exist?’ The same plea is still urged by the
defenders of Zola’s minute details in descrip-
tion of misery, profligacy and filth. The
discussion is often closed by the argument
that it is the province of Art not to teach,
but to please. This would not end the
debate between Euripides and Aeschylus,
for they agree that it is the duty of Art ‘ to
make men better.’
The reader will receive with interest the
judgmenton this question of the distinguished
scholar to whom we owe the present treatise:
2 Frag. 170 (Nauck).
3 ἄμαχος yap ἐμπαίζει θεὸς ᾿Αφροδίτα, Soph. Ant.
4 οὐδ᾽ οἶδ᾽ οὐδεὶς ἥντιν᾽ ἐρῶσαν πώποτ᾽ ἐποίησα γυν-
αἴκα, Ran. 1044.
I
114
we commend it to M. Zola: ‘le public a
plus de plaisir aux fictions qui l’élévent, qu’
aux réalités qui le degradent. L’Art, sans
TWO EDITIONS
Andocidis Orationes edidit Tustus HERMANN
Liesius; pp. xxxii, 67. Β. Tauchnitz,
Leipzig, 1888. Μ. 1. 20.
Tur interest felt in the announcement that
an edition of Andocides was in preparation
by the reviser of Der Attische Process was
somewhat tempered by the knowledge that
the work would be purely critical. The
fact that no satisfactory explanatory edition
of this author had ever been written mde
it all the more desirable that the entire field
should first be covered by a scholar who had
given proofs of his competence to deal with
the general subject.
The editor however gives a very valuable
introduction on the life and writings of An-
docides, where in copious notes the reader is
referred to the sources, ancieut and modern,
upon which he bases his judyments. His ar-
gument against the genuineness of the so-
called Fourth Oration is a most powerful one,
and further proof will scarcely be needed that
it is the work of a late rhetorician. His
belief that the pseudo-Lysian oration against
Andocides was a similar exercise, and not a
contemporary argument in court, will hardly
yet find such general acceptance. He
promises an article on the subject, but Lam
not aware that it has appeared. Upon the
critical apparatus of this edition it is not
necessary to enlarge here, although it is dis-
tinctly the most valuable part of the book.
The spelling is based throughout upon the
combined resu.ts gleaned from Attic con-
temporary inscriptions which are certainly
better authority for spelling than the most
careful copyist of the Middle Ages, and we
must accordingly expect to find the latter
give place in our newer texts of the authors.
Such emendations as are due to the editor,
except mere verbal corrections based upon
the new collations of MSS., havein the main
already appeared in his contributions to
German periodical literature, and are there-
fore known to readers. But we are given
also a collation of the suggestions and cor-
rections of other scholars, most conscienti-
ously gathered and destined to be of great
helpfulness. On the whole, this edition, in
THE CLASSICAL REVIEW.
cesser d’avoir le plaisir pour but, le cherchera
plutdt dans la transfiguration que dans la
copie de la réalité.’
R. Y. TyRReE.L.
OF ANDOCIDES.
spite of a number of emendations which
have elsewhere met with criticism, may be
safely recommended as the most useful
to the critical student which has ever
appeared.
Andocidis de Mysteriis et de Reditu ; edited
by E. C. Marcuant, B.A., late scholar of
Peterhouse, Cambridge; Assistant Master
at St. Paul’s School. Rivingtons, London,
1889. 5s.
Mr. Marcuant’s text, also, is preceded by
an essay on the life of Andocides and by a
full analysis of the two orations. The his-
torical circumstances are ably treated with
fair references to authorities. At the end
of the book there is an appendix on the
connexion of Andocides with the Hermes
affair. In this, Mr. Marchant concludes
that Andocides had fully intended to take
part in the business. So far he is probably
right, but what about the reasons he assigns _
for the inaction of Andocides ἢ ‘ His absence
was due to scruples at the last minute, or to
the fear of detection.’ To believe this, it is
necessary to discredit the whole of Andocides’
description of the affair as given in Myst. ὃ
61 sq. Mr. Marchant thinks that he is justi-
fied in disbelieving all the statements in the
later oration by the admissions made in the
earlier (de Reditu, § 7), which he interprets
as a declaration of guilt ; he avoids stating,
however, whether he believes that Andocides
was really an actor in the crime or not. I
think that the statements in the two orations
may be partly, although not wholly, recon-
ciled. We are not obliged to believe from
de Red. § 7 that the speaker is confexsing
participation in the actual mutilation of the
Hermae. Perhaps he means only to say
that he was in the plot and had intended to
take part in its execution. He was thus in
a position to inform against the actual
criminals and at the same time to clear
himself from the act of crime, although he
does not try to avoid the charge of being to
some extent implicated in it. He doubtless
expected complete indemnity, and indeed
THE CLASSICAL REVIEW.
got it for a time, but the decree of Isotimides
was sweeping enough to cover his case, and
he went into exile. The story about the
fall from his horse is too natural and too
easily refutable to be a lie; not so his
statement that he refused altogether to
share the plot. This may fairly be dis-
credited, and one may think that Euphiletos
was ignorant of this accident and expected
Andocides to do his part. I should there-
fore reject only the latter’s statement that he
refused to join the plot and that Euphiletos
knew it. His absence, then, on the event-
ful night would be due to his accident, and
not to scruples, least of all, in such a man,
to fear. ‘This explanation may reconcile
perhaps the statement of Thucydides (vi. 60)
that Andocides gave evidence against him-
self. However, I do not feel as anxious as
some commentators over any apparent con-
tradiction of Andocides by Thucydides. The
historian, it will be remembered, left Athens
eight years before the Hermes affair and
could not have returned till eleven years
after it. In so complicated a matter it
would be difficult for any one, not actually
on the ground, to know the exact details.
Thucydides himself admits that nobody had
ever been able to be sure about them. It
is rather surprising that Mr. Marchant, who
seems to realize (pp. 127 and 136) the
untrustworthiness of Thucydides in two
particulars, should elsewhere (pp. 35 and
178) accept his evidence with little question.
Mr. Marchant is perhaps nearer the truth
in other places (pp. 11 and 22) and when
he says (p. 6) ‘Andocides gave evidence
incriminating himself and others,’ but even
‘incriminating’ appears to me scarcely the
proper word in this connexion. But after
all that has been written on this vexed sub-
ject, we are as far as ever from the last
word,
Mr, Marchant is familiar with the work of
Lipsius along the lines of legal antiquities,
and treats fully of legal matters as far as
they fall within his subject. Perhaps
teachers would have been better satisfied
with fuller references to the sources. It
is rather unfortunate that all the references
to Lipsius’ Attische Process should be by
sections, the fact being that Mr. Marchant
has taken for sections the paye-numbers of
the old edition found on the margin of the
new. This is frequently confusing, especially
on p. 155, where the reference should read
Att. Proc. p. 900, note 383. The page-num-
bers of Stephanus are omitted. They should
always be given as long as Liddell and Scott
continue to quote by them.
115
Not much grammatical comment is at-
tempted, but what appears is as a rule
founded upon acquaintance with the general
subject and with the habits of this particu-
lar orator. Ina very useful note on Myst.
$10 the reference to Kiihner should read
423 C. On ὃ 30 we are given some sound
doctrine on δῆλον ὅτι. Just above, in the
same section, Mr. Marchant reads with
Blass ἡγήσεσθε, where Lipsius has allowed
the unnecessary ἡγήσαισθες In his note on
δ΄ 21 the proper reference to Professor
Goodwin’s M. and. 7. is § 49, 2, N 3, (ὁ),
although § 44, 3 may also be compared. In
the de Red. ὃ 16 my feeling is that the
notion of possibility, so prominent in ὅδόν τε
καὶ πόρον, takes away any ‘harshness’ from
the following infinitive. In his translations
the editor is correct enough, bnt not always
idiomatic. There is here and there the
especial weakness of rendering a long Greek
sentence, composed of many subordinate
clauses, into a correspondingly long one in
English (cf. p. 126).
The comments on two passages may be
mistaken. In Myst. § 41 Leagoras was not
represented as supposing that Dioclides had
come to ‘join the plot,’ but to listen to pro-
posals of the conspirators that he should
keep quiet about it. In § 132 we read ‘of
course μυῶν does not mean A. himself per-
formed the rites, but he paid the expenses.’
The first part of this is right, for A. was
nothing but μυσταγωγός, but what were the
expenses? The μυσταγωγός, as 1 know him,
merely presented the candidate, and, himself
an initiate, took charge of the latter as a sort
of ‘Mentor’ during his initiation. Do we
know anything of initiation fees? Mr.
Marchant wisely refrains from saying much
about so dark a subject as the Mysteries ;
he might well have omitted his quotation on
δ 31 from C. R. Kennedy. ‘Solemn revela-
tions’ looks too much like a reference to a
fancied expounding of doctrines. Lobeck
exploded the notion that any preaching was
done at Eleusis.
In the matter of spelling the editor
wisely follows Lipsius, and goes even further
in the right path when he prints τοῖν for
ταῖν (Myst. § 144). We have long been
warned in grammars of the feminine dual of
the article ; it utterly disappeared from F.
D. Allen’s edition of Hadley, but it still
survives in a few passages in the authors as
printed.
In the critical notes the editor’s system
seems to have been to present only the colla-
tion of those passages wherein Blass and
Lipsius differ, and to state which of them he
12
Ε10
himself has followed. In 77]γ8ὲ. δ 4 the emen-
dation attributed to Lipsius was made by
Képke in 1864 except for the spelling duped.
Realizing the danger of emending according
to stereotyped rules so easy-going a writer
as Andocides, the editor is very conservative.
His restoration of the MS. reading de Red. ὃ
22 seems sound, and also his transposition of
χρήσασθαι in Myst. § 86. The correction of
ταῖν χεροῖν (Myst. ὃ 144, by a misprint
referred on p. iv. to§ 141) has already been
mentioned. In § 153 his change to the aorist
seems unnecessary, although it might be
accepted in the case of a more exact writer.
In ὃ 12, for τὸ μειράκιον ὃ Mr. Marchant
reads ὃν, on the ground that an Attic ear
would not bear the neuter. It is true that
the natural gender prevails in such cases,
yet we have a neuter with μειράκιον in Anti-
THE CLASSICAL REVIEW.
phon Zetral. By, ὃ 10, and in Plat. Prot.
315 D, side by side with the masculine ; cf.
315 ἘΞ Neither of these examples is an
exact parallel, but after all who shall decide
on the nicety of the ear of Andocides ?
The remarks here made are by no means
written with a wish to detract from the just
dues of so meritorious a book as the one
before us. Every one must be grateful to
Mr. Marchant for giving English and
American readers an excellent practical
edition of an author who has been too long
neglected ; for his fairness in stating both
side of disputed questions, his clearness in
summing up results, and the modesty with
which he differs from the more dogmatic
utterances of some of his predecessors,
Morris H. Moraay.
Harvard University.
COLERIDGE’S TRANSLATION OF APOLLONIUS RHODIUS.
The Argonautica of Apollonius Rhodius
translated into English prose from the text
of R. Merxert by Epwarp P. CoLeripGE
B.A. George Bell & Sons, 1889. 5s.
THERE are in English three poetical or rather
verse translations of Apoll. Rhod. by
Fawkes, Greene, and Preston respectively,
more or less near the original, but it is
certainly time that there should be a prose
translation, and so the present one is wel-
come. The interest of the poem at the
present day is mainly literary, not to say
philological, and apart from the Greek the
story of the Golden Fleece as told by
Apollonius will not attract many readers.
The present translation though it cannot
be compared to the best of Bohn’s, such as
Dale’s Thucydides or Kennedy’s Demosthenes,
is yet above the average and may be con-
sidered adequate. At the same time it
bears many marks of haste and some of the
errors I shall presently point out seem to
be due solely to that cause. ‘The translation
is from the Teubner text of 1852 based on
the Laurentian MS. In the next year came
out Merkel’s larger edition which contains
many improvements over the ed. min. Mr.
Coleridge apparently adopts the theory that
Apollonius should be translated in an archaic
style. J have nothing to say against such a
theory, indeed much may be said for it, but
it is not carried out by writing in a modern
style and interspersing archaic words and
phrases. The archaism should lie rather in
the structure of the sentences than in single
words. Thus we find such scattered gems as
‘astonied,’ ‘God wot,’ ‘I wis,’ ‘swinked,’
‘an’ (=if), ‘took up his parable’ ( = replied)
‘or ever,’ ‘ yestreen,’ &c. ‘ Unbeknown to thy
father’ may be archaic, but its associations
are rather with Mrs. Gamp, and ‘for to.
behold’ is a vulgarism now at any rate.
Mr. Coleridge has a sufficient vocabulary at
his command and can use it with effect, e.g.
TV. 152 ‘like a dark wave dumb and noise-
less rolling o’er a sluggish sea.’ Alliteration
is no stranger to him, as ‘a flash as of flame
from the flashing of the fleece.’ The passage
of the Symplegades II. 551 foll. is very
well rendered on the whole. It is rarely we
find such an awkward sentence as ‘ what
time, alone and apart from the other chiefs,
he routed them’ &e., but why will he always
write Lolchos 1?
The translation is prefaced by a short
notice of the life of Apollonius in which the
date of his birth is given as ‘about B.C.
235.’ I should have thought this to be a
misprint, if it were not that Smith’s Dic-
tionary of Biography gives the same date.
Though the precise year of Apollonius’ birth
is uncertain, the limits of deviation are
narrow and Ritschl’s date of Ol. 126 (276-273
B.c.) is probably correct. I will now notice
more particularly some points in the trans-
lation and notes which require correction.
I. 438 ‘Now on that day were his limbs
THE CLASSICAL REVIEW.
weighed down with wine.’ There is nothing
about wine in the Greek, which is τότ᾽ αὖ
βαρύθεσκέ οἱ ἤδη | γυῖα and means of course
‘heavy with age,’ and then comes the usual
antithesis between age of body and youth-
fulness of spirit. 368 ἔζωσαν πάμπρωτον
ἐυστρεφεῖ ἔνδοθεν ὅπλῳ ‘they lashed the ship
stoutly with a well-twisted cable from
within.’ ‘From within’ is nonsense. The
ὑπόζωμα is meant and we should probably
read ἔκτοθεν : at any rate some emendation is
necessary. 1110 ἤρεσαν és λιμένα Θρηίκιον (note)
‘It isclear that the heroes did not sail across
to Thrace because Mt. Dindymus is a con-
siderable distance inland in Galatia, and it
was hither they meant to come.’ The Mt.
Dindymus here meant is that at Cyzicus, as
is clear from 985 sup. 1277 ὕψι δὲ νηὸς |
εὐναίας ἐρύσαντες ἀνεκρόυσαντο κάλωας ‘and
they hauled in the anchor ropes of the ship
and backed her out.’ εὐναίας is a noun =
εὐνάς, and if it were an adj. could not agree
with κάλωας masc. Without some emenda-
tion, such as εὐναίους, it is not easy to see
the meaning of ἀν. κάλ. Schol. Laur. ex-
plains προσέκρουσαν τῇ νηὶ τὰ σχοινία τῆς
ἀγκύρας διὰ τὸ βεβρέχθαι. In a note it is
said ‘The ancient mode of landing was to
beach the shipif possible and then fasten by
cables to the land, and by anchors from the
stern in the sea.’ This is the exact opposite
of the fact. The Homeric mode of landing,
as ‘every schoolboy’ knows, is to fasten the
stern by cables (πρυμνήσια) to the land, and
to steady the ship by large stones (civa/)
used as anchors, cast out from the prow, cf.
A 436, +137 and Apoll. Rhod. IT. 160. II.
163 (note) ‘The Therapnaean son of Zeus,
z.e. Apollo, so called from Therapnae, a part
of Sparta, which was sacred to this god.’
So Schol., but more probably Polydeuces,
is meant who had just conquered Amycus,
Therapnae being the birthplace of the
Dioscuri. 592 ὅσσον δ᾽ ἂν ὑπείκαθε νηῦς
ἐρέτῃσιν, δὶς τόσον ἂψ ἀπόρουσεν. ‘So she
sprang forward twice as far as any other
ship would have yielded to rowers’ (!),
literally ‘but as far as the ship yielded to
the rowers it leapt backwards twice as far’
1.6. through the force of the waves. The
difficulty is to assign any meaning to ἄν:
hence some (Brunck and Beck) read παρείκαθε
with an inferior MS. 1064 (note) ‘ ἄρσετε
Tonic fut. from ἀραρίσκω ᾿ : more probably aor.
imper. like Homeric οἴσετε. III. 1 (note)
“Ἐρατώ the Muse of Dancing...As this third
book is to relate Jason’s wooing and winning
of Medea, there is a certain appropriateness
in an address to the Muse who presided over
such festivities as were customary at wed-
ἘΠῚ
dings.” The Muses were invoked quite in-
differently by the poets, as eg. by Horace,
and the introduction of Erato is clearly due
to the play on ἔρως : see l. 5 τῷ καί τοι ἐπήρατον
οὔνομ᾽ ἀνῆπται, Which is imitated by Ovid (Am.
11. 16) Nune Erato, nam tu nomen amoris
habes. Moreover, if we assign separate
functions to the Muses, Erato is not the
Muse of Dancing but of Erotic poetry and
so strictly in place here. 206 ἀλλ᾽ ἐν ἀδεψ-
ἤτοισι κατειλύσαντε βοείαις κιτιλ. ‘but two
men must roll them up in hides untanned’
etc. and note ‘it is difficult to believe that
Apollonius wrote the dual.’ It is probable
that Ap. here uses the dual for the plural in
imitation of Zenodotus, so in I. 384. 613
‘lest haply he should win her over in vain...
or lest if she consented to his prayer’ ete.
The subject of μειλίξαιτο is not Argos but
Chalciope ; ἀτυζομένην and ἑσπομένης both
refer to Medea. 988 ὑμείων cannot mean
merely ‘thee,’ but alludes to Hecate and
Zeus as well as to Medea. 1018 τῆς δ᾽ ἀμαρυγὰς
ὀφθαλμών ἥρπαζεν ‘snatched bright glances
from her eyes’ : rather ‘ snatched (¢.e. captiv-
ated) her sparkling eyes,’ cf. rapere ocellos.
1025 φράζεο viv ὥς κέν τοι κιτιλ. not ‘ that Τ
may devise’ but ‘how I shall devise.’ 1058
‘but be not thyself eager for the fray,’ strike
out not which is probably a slip. 1092
(note) ‘The isle of Aea was a small island
in the river Phasis in which the golden fleece
was kept.’ Not at all: the isle of Aea here
meant 1s the abode of Circe in the West, as is
clear from sup. 1073. 1104 συνημοσύνας
ἀλεγύνειν ‘to heed the ties of kin.’ No: ‘to
heed compacts (or agreements).’ 1116
ἀπροφάτως means not ‘openly’ but ‘ unex-
pectedly.” 1118 ὑποβλήδην προσέειπεν not
‘caught her up and said’ but ‘said in
answer.’ This is always the meaning of ὑπ.
in Apollonius. 1137 ἐδεύετο δ᾽ ἤματος ὥρη |
ay οἷκόνδε νέεσθαι κιτ.λ. ‘for the time of day
demanded the maiden’s return home’ ete.
This cannot be correct : édévero is here used
absolutely as in Y 122, lit. ‘ the time of day
was wanting (or failing) for her toreturn’ ete.
1285 παρὰ δ᾽ ὄβριμον “ἔγχος ἔπηξεν Ι ὀρθὸν
ἐπ᾽ οὐριάχῳ ‘and fixed his strong sword up-
right to the hilt hard by.’ ἔγχος is spear not
sword-—in IV 223 it is again mistranslated
sword—and ἐπ οὐριάχῳ means ‘upon the
butt-end’ which was furnished with a spike.
1298 (note) ‘the ydavos is the mould into
which the liquid metal is poured for casting.’
The x. is the smelting-furnace, or perhaps
melting-pot, as it is translated in the text.
1334 λαῖον ἐπὶ στιβαρῷ πιέσας ποδί “ pressing
down the left stilt with heavy heel.’ This is
nonsense. λαῖον 15 a rare word meaning
118
ploughshare. Jason presses his foot on it to
make it go deeper into the ground as plough-
men often do nowadays. 1385 ἀμφ᾽ οὔροισιν
ἐγειρομένου πολέμοιο not ‘upon his boundaries’
but ‘concerning boundaries,’ which were a
frequent subject of strife. IV. 16 τάρβει
means not ‘terrified’ but ‘was afraid of.’
24 ‘poured from the casket all her drugs at
once into the folds of her bosom.’ No one
would think that the Greek describes exactly
the opposite action, viz. the pouring of the
drugs from her bosom into the chest. 78
ἠπείροιο not ‘mainland’ but ‘shore,’ as often.
304 Πόντοιο not ‘into Pontus’ but ‘ out of
Pontus,’ probably a slip. 337 μέσφα with
gen. not ‘between’ but ‘as far as to’ and
Νέστιδος αἴης, which is described in the note
as meaning Thrace from the river Nestus,
really means a part of Illyria and has no
connexion with the Thracian river of that
name, cf. inf. 1213 Neoratovs. 379 pe μάλ᾽
εὐκλειής ; ‘will not my fame be passing fair 4’
dele ‘not.’ 436 θελγέμεν, εὖτ᾽ ἂν πρῶτα θεᾶς
THE CLASSICAL REVIEW.
περὶ νηὸν ἵκηται | συνθεσίῃ... | ἐλθέμεν “50 as
to persuade them to depart, as soon as
Absyrtus came by agreement’ &c. So Wellauer,
but rather ‘to persuade him (Absyrtus) to
come as soon as she (Medea) came’ ἄς. 501
ῥηιδίη δέ Kev ἄμμι... | ἤ τ᾽ (or ἥ δ᾽) εἴη μετέπειτα
κατερχομένοισι κέλευθος *’twill be an easy
route for us or indeed for any who come
hither hereafter.’ Not so; with 7 δέ, κατ.
refers back to ἄμμι, with 77 to the Colchians.
530 περόωντι κατὰ χρέος not ‘in obedience to
an oracle’ but, as Beck translates, ‘ negotii
sui causa.’ 552 περιώσια σήματα νηός ‘ Argo’s
wondrous pennon’ probably means figure-
head or some other ornament, see inf. 1618.
578 (note) ‘ Hera brought them by contrary
winds to the island of Electra, in order that
Jason and Medea might there be purified by
Circe.’ Not the island of Electra, but the
island or rather promontory of Circe is where
they were purified, inf. 659. 705 δέρην
is translated ‘skin’ as if it were δορήν.
R. C. Seaton.
OWEN’S EDITION OF THE 7RISTIA.
P. Ovidi Nasonis Tristium Libri V.
Recensuit 8. G. Owen. Oxford. Claren-
don Press. 1889. pp. cxili+ 271. 16s.
ΤΉΟΒΕ who, after reading Mr. Owen’s small
edition of the first book of the Zristia pub-
lished a few years back, looked for some
more elaborate work on Ovid from his pen,
will certainly not be disappointed in the
volume before us. His new book is a model
of sound criticism, laborious research and
orderly arrangement; and in every way a
real honour to English scholarship. Mr.
Owen has collated nearly thirty manuscripts
of the Z’ristia, and has consulted or secured
collations of several others. He also gives
the readings of seven deflorationes and tour-
teen old editions. So we are furnished with
a splendid apparatus criticus. Any one who
has studied the text of Ovid with the help
of the meagre MSS. evidence given in such
respectable editions as those of Riese, Korn,
Zingerle or Ehwald, must sincerely hope
that our editor will do similar service for
other parts of his author.
The book opens with a very complete
introduction in eight sections, written in
Latin, which, though sometimes too verbose,
is always intelligible. The chapter headed
de codicum rationibus is an interesting and
solid contribution to textual criticism. By
a minute and ingenious process the MSS. are
compared and divided into families. Mr.
Owen shows that six of his MSS. (in-”
cluding the Marcianus, to which of course
he gives the supremacy) are to be traced to
one archetype. He also establishes beyond
reasonable doubt his coutention that the
codices deteriores form quite a distinct family.
Hence he gets a satisfactory basis on which
to work. Another valuable section of the
introduction is that entitled guibus temporibus
Trisiia recensa sint, in which it is argued
chiefly from the evidence of the MSs. that
the Zristia must have undergone several
recensions, one before the fourth century and
two or three others in the middle ages.
There is one thing in this introduction
against which I should like to protest, and
that is the Latinization of the names of
modern Englishmen, which often produces a
grotesque result: eg. Coxius, Robius,
Nettleshipius, Shuckburghius, [Thackerayus,
Wilkinsius. Surely these might have been
left as indeclinables.
Mr. Owen has adopted the wise plan of
consigning to an appendix the bulk of the
conjectures made on the Zristia by older
scholars, mentioning in his critical notes
hardly any besides those which he -admits
THE CLASSICAL REVIEW.
into the text. This long list of emenda-
tions (among which we tind Madvig’s sz
iam deficiam suppressaque uena paletur,
111. 3, 21) is a sad monument of wasted
ingenuity. To this list Mr. Owen has added
but little, for he is a model of caution in
editing. His improvements in the text are
mostly due, not to brilliant divination, but to
the wise restitution of the readings of good
MSS. As an instance of such restitution
we may quote II. 77—80 as it stands in his
text :—
a! ferus et nobis crudelior omnibus hostis,
delicias legit qui tibi cumque meas,
carmina ne nostris quae te uenerantia libris
indicio possint candidiore legi.
Other good examples are II. 378 corpora
pressa toro, 11. 277 at quasdam witio, 111. 4,
73 quamuis longe, IV. 1, 29 wis me tenet ipsa
sacrorum, 1V. 4, 62 friyoris axe, IV. 9, 3
dementia, V. 1, 23—24 animos—met, V.1, 66
sic mihi, V. 6, 5 si non tempore nostro. These
readings are ably defended in the chapter
entitled Vindiciae ; and in some cases new
and ingenious interpretations are offered.
The few emendations of his own which
the editor introduces into the text, though
they are not all convincing, are at least
improvements on the vulgate. The most
noteworthy are 11. 85 cum quae, 11. 479 ut
comitare sequens, ILI. 3, 21 deficiat sub crasso
lingua palato, IV. 1, 104 iste, V. 10, 12
iempora uersa, V. 13, 6 scilicet inmoddico.
In IV. 10, 96 Mr. Owen prints a clever
conjecture communicated to him by Prof.
Strachan, equis for eques. The dithcult
passage 11. 191—198 is greatly improved by
a rearrangement of lines and emendation of
the proper names.
On a few passages I venture to offer some
small criticisms :—
I. 3, 52 uel quo festinas 176. uel unde, wide.
The MSS. are certainly in favour of festinas.
But it is hard to believe that an Augustan
poet could have used the indicative in such
acase. Almost all Mr. Owen’s parallels are
instances of sunt gui with the indicative, a
well-known Ovidian use and surely far less
curious than quo festinas uide. We cannot
argue from such passages that Ovid is
generally loose in his use of moods. Mr.
Owen refers also to Metam. X. 637. Does
he really believe that Ovid wrote there guid
Sacit ignorans ?
11. 232 pars nulla est, quae labat, imperit.
The Marcianus has lJabat; but almost all
other MSS. have dabet, and the variants point
119
to the same reading. It seems to me that
this is just one of those passages where the
authority of the Marcianus is outweighed
by the consensus of the other MSS. Besides,
quae labat after a preceding negative is
strange syntax indeed, which cannot, I think,
be supported by Mr. Owen’s instances
referred to above.
IV. 10, 7 non modo fortunae munere factus
eques. Though I quite agree that Mr. Owen
gets the right meaning in this and the pre-
ceding line, 1 cannot think with him that
Jortuna here = res familiaris. Fortunae
munere must mean ‘by the bounty of
Fortune.’
V.1, 71 ipse nec emendo, sed, ut hic de-
ducta legantur,
non sunt illa suo barbariora
loco.
Mr. Owen’s interpretation is, I venture to
think, rather clumsy. Perhaps we should
explain ut—legantur as meaning ‘though
what is read (at Rome) has been composed
here (in Scythia),’ putting the stress of the
clause on the participle. But the passage 15
a very doubtful one.
V. 10, 41 utque fit, in se aliquid siatui, αὐ.
centibus wlis
abnuerim quotiens annuerim-
que, putant.
in se...staiui is a good emendation. For the
subjunctives in the second line Mr. Owen
quotes Cic. Cluent. § 86 quodcwmque diceret,
honestius diceret. Surely they are due to
Oratio Obliqua: ‘they think that, as often
as I have nodded...’
With regard to the vexed question of
Latin orthography Mr. Owen has no pre-
conceived system, but, as a rule, follows the
MSS., except where their spelling is clearly
incorrect. Purists in spelling will be sur-
prised to find that he prints welgus and
uultus with the MSS., in fact ww, not wo,
always ; and that he makes out a good case
for anchora (p. xeviii). For brachia (so the
MSS., not bracchia) he refers to Riese, vol. I,
Ρ. xiii. But Riese’s argument seems to me
absurd. Mr. Owen writes Laodamia. 1
thought that the evidence was rather in
favour of Laudamia.
The excellent collotype facsimiles of pages
of the Marcian and Tours MSS. add greatly
to the value of the work. They will be of
much service to students of textual criticism.
It is perhaps hardly necessary to add that
the whole book is got up in that admirable
style which we have learnt to expect from
the Clarendon Press.
G. M. Epwarps.
THE CLASSICAL REVIEW.
HARDY’S EDITION OF PLINY’S CORRESPONDENCE WITH TRAJAN.
CO. Plinii Caecilii Secundi epistulae ad Traia-
num imperatorem cum eiusdem responsis.
Edited, with notes and introductory es-
says by E. G. Harpy, M.A. London. Mac-
millan. 1889. 8vo. pp. xii, 251. 10s. 6d.
Mr. Harpy is to be congratulated on the
choice of his subject, both in this volume
and in the edition which he promises of Plu-
tarch’s lives of Otho and Galba.
He has been fortunate enough (see his
article in the Journal of Philology, vol. xvii)
to discover in the Bodleian a volume con-
taining the editions of Beroaldus (1498) and
Avantius (1503, part of the correspondence
with Trajan) together with a transcript of
the letters wanting in both collections. It
seems certain that this Bodleian volume was
used by Aldus as copy for the first Aldine
edition (1508). Even Keil’s larger edition
thus receives a welcome supplement from
Mr. Hardy’s collations.
The mass of new material brought to light
chiefly from inscriptions, and digested by
Mommsen and Marquardt (Mr. Hardy I
think always, certainly seventeen times,
writes Marquadt ; if any consonant in this
clumsy name is superfluous, surely it is the
d or t), suggested to Mr. Hardy the illustra-
tion of the lUitterae inlitteratissimae (he
writes (p. vi) a single ¢ in each word ; p. 29,
where he gives the reference, has the right
spelling) of the official correspondence between
Trajan and Pliny. It is a great convenience
to have collected at the foot of the page
what must else be sought in several costly
volumes. Even possessors of the best old
editions will be grateful for the saving of
time and labour which this work offers to
them.
Mr. Hardy complains that Pliny’s letters
to Trajan ‘have hitherto received compara-
tively scanty notice from editors and scho-
lars.’ It would have been well for the
solidity of his undertaking, if he had used
the labours of such pioneers as exist. He
betrays no acquaintance with the elaborate
edition of Gottlieb Cortius and P. D. Longo-
lius (Amst. 1734, 4to), with its index of
words and the valuable elucidations (for the
so-called Tenth Book) of Rittershusius, Buch-
ner, Longolius : in this work, and in that of
G. E. Gierig (Leipz. 1802 8vo.) almost all
difficulties, of matter or of language, are
fully discussed by masters of Silver Age
Latinity. If to these two editions, and the
slighter work of J. M. Gesner, as edited by
G. H. Schiifer, he had added the colossal
commentary of Schwarz on the panegyric
and Lagergren’s dissertation on the life and
language of Pliny, he might safely have
neglected the German notes of Moritz (not
Moritz, as we read p. 74) Doring (Freyberg
1843), which have received undue attention
from him and from other English editors.
Any one who will verify Déring’s citations
through a few letters, will learn to distrust
his guidance.
For the famous letters on the Christians,
Mr. Hardy’s choice of authorities is especi-
ally unfortunate. ‘For my notes and es-
say on the Christians, I have consulted the
dissertations of Bandouin (read Baudowin)
and Vossius ; Aubé, Histoire des Persécutions ;
and Schiller’s Geschichte der Kaiserzett.’
Schiller and Aubé are notoriously weak on
the side of ecclesiastical history; a few
hours’ study of the sober critics Kortholdt in
the 17th century, Boehmer, Moshein, Lard-
ner in the 18th, would have enabled our
editor to form a clearer and more consistent
view of the bearing of Trajan’s rescript than
he conveys to the reader who compares his
scattered utterances on the subject.
On p. 12 we read: ‘This reply made Christianity
a definite political offence, and laid down the prece-
dent which in fact characterised the subsequent per-
secutions.’
P. 62: ‘Trajan commends Pliny’s past procedure,
and distinctly lays it down that all who are charged
by responsible accusers and prove to be Christians, on
refusing to deny the name, and, asa test of bona fides,
to worship the state gods, are to be executed. This,
then, is the first distinct and legal ordinance made
respecting the Christians. It amounts, there is no
doubt, to an edict of proscription. The profession of
Christianity, as such, if proved on satisfactory
grounds, was a punishable offence....That the edict
opened out the possibility of a legal persecution there
is no doubt, for it was quite within the competence
of the governor to invite accusations though he could
not initiate them.’
p. 63: ‘With regard to the application of the
edict, I think it is quite clear that it related not
to the empire as a whole, but to Bithynia only.
Trajan’s own words, ‘neque in universum aliquid...
constitui potest,’ partly imply this, but apart from
this, without definite evidence, it would be quite in-
admissible to suppose that a rescript given to a par-
ticular governor in answer to particular questions
could be applied beyond the province about which it
was written.’ [Yet the corpus cwris and Haenel’s
collection have no meaning except by virtue of this
supposition. ]
ibid. ‘Trajan’s rescript is really little more than a
supplement to his previous edict concerning the
THE CLASSICAL REVIEW. 121
hetacriae in Bithynia, and really did not touch
Christianity as a religion at all.’ cf. p. 248.
It is melancholy, in the nineteenth cen-
tury, to find on epp. 96 97 the loose cita-
tions of Catanaeus or Baudouin reproduced
without attempt at verification. Thus ‘ Ul-
pian’ (p. 211), ‘Hieronymus in Chron.
Euseb.’, ‘Tertullian’ (p. 212), ‘ Prudent.
passione Vincenti,’ ‘Tertullian’ (p. 213),
‘Tertullian,’ ‘ Euseb. Zecl. Hist. V,’ ‘ losephus
Ant. Iud. 14,’ ‘Philo,’ ‘Dio Cass.’ (p. 214),
‘Ulpian lib. viii, de officio Procons. cap. 1,
‘again,’ ‘in the persecution in Gaul under
Marcus Aurelius it is said,’ ‘Tertullian to
Scapula,’ ‘Doring appositely quotes Pru-
dentius’ (p. 215, Doring of course had not
seen Prudentius, whose words are given by
Longolius with the precise reference).
‘Tertullian,’ ‘ Eusebius,’ ‘Cic. pro Roscio’
(p. 216). ‘Tertullian ad Scapulam, Tertull.
an Scep. (sic, p. 217).
The misprints, especially in proper names,
in German and in Greek, are very frequent.
I count the barbarism connection ten times ;
but, until correctors and critics unite in
proscribing the οὐ in substantive derivatives
of necto and flecto (which to a scholarly eye
are aS nauseous as resurrexion), we cannot
visit a universal error on any single head.
But no such excuse can be made for the
ingenuity which rolls into one the two editors
John Scheffer (1621-1679), librarian to
Queen Christina of Sweden, and George
Henry Schiifer, best known here as editor of
Porson’s four plays and of Reiske’s Demos-
thenes. Preserving the ff of the earlier,
and the ae of the later name, Mr. Hardy, to
avoid favoritism, gives us for both (once, p.
74, within the space of four lines) Schaeffer.
Wilmanns, the epigraphist, when his name
is not abridged, appears as Wilmann (pp. 48,
75, 79a, 85b, 95b, 109b, 1118, 115ab, 138a,
2510, 2998) or Wilman (2020 bis, 2090,
2220). The abbreviation rim. (for rémische)
always figures as rom. I give a specimen of
other errors as they occur.
p. 8n. 1 Beitrage pl, Trajan gen. p. 15 1. 4 from
foot ‘ Henri Francke.’ Why not Henry (or Heinrich,
as it stands in the book)? p. 311. 9. from foot quia
(1. quin). p. 56 1. 8 from foot κἀν (=xal ἐν), in
defiance of Porson’s warning. p. 78a]. 10 from foot
Antonius’ (1. Antoninus). p. 79a ‘succurit’ and
(for arrogantia) ‘arroganti.’ yp. 87a ‘for a son to re-
fuse the hereditas was most dishonourable ; see Cic.
Phil. 11 16, quamquam hoc maxime admiratus sum
mentionem te hereditatum ausum esse facere, cum
ipse hereditatem patris non adisses.’
with a shilling’ cannot be said to refuse his patri-
mony. p. 88 on letter 5 1 miss a reference to letters
6 and 7 and 10, which relate to the same matter.
pp- 89 (text) and 90a (note) the barbarous form
conditionis, though Keil has condicionis. p. 90a
A son ‘cut off
Theomuthis, but p. 89 rightly ‘Thermuthin.’ p. 92
a (summary of ep. 7) ‘home’ for ‘nome.’ p. 920
three false accents. p. 94b ‘ Quintitius’ for ‘ Quinti-
lius.’ p. 97a bis ‘vindematio.’ p. 100 title to ep.
11 ‘media’ for ‘medici.’ p. 1180 ‘Gierig cites.’
Not Gierig, but Déring in reply to Gierig. p. 116
on letter 23 should be cited 70, on the same subject.
p. 118a an impossible accent αἴρειται. p. 1190 three
monsters Tato’. ταμιάς. tpds (for mpds). p. 1910
‘numerals enclosed by two vertical and one horizon-
tal line are multiplied by 100,000, and those by a
single horizontal line, by 1000.’ A fine example of
zeugmna. p. 160b ἕν. yp. 165a ‘atrociosa.’ p. 202b
κεχιλιαρχήκως. ἔλπιδα. p. 2120 ‘exprobatis.’ ‘ex-
probavit.’ p. 214b Tertullian [add ‘de ieiun. 17’)...
implies that abuses crept in, ‘ Apud te agape in sae-
culis [1]. caceabis] fervet: fides in culinis calet: spes
in ferculis iacet.’ The context ought to have guarded
our editor from this slip. p. 216b ἔμπεσον. p. 221b
‘ perigrini.’ ‘ milariae.’ p. 221b ‘ vitem poscet libello,’
for ‘ posce.’ p. 230 (text) ‘debeant’ ; (note) ‘I have
adopted with the substitution of debcrent for possent.’
p- 231b ‘Suet. Claud. 32 convivia agitant []. agitavit]
et ampla et assidua ac fere patentissimis locis, ut
plerumque sexcenteni simul discernerent ’ []. disewm-
berent].
A great boon would be conferred on
readers of Pliny, if the wish of Fabricius
were carried into effect, and the commentary
on the epistles, left ready for press by one of
the most learned of English Latinists, John
Price, were given to the world. A lexicon
to Pliny, to match Bonnell’s to Quintilian
and Gerber and Greef’s to Tacitus, is also
greatly needed. It would, for one thing,
make it impossible for any sane man to
question the genuineness of the letters re-
lating to the Christians. The following
parallels will suffice to shew how absolutely
identical they are in style with the remainder
of the book.
ep. 90 δ 1 cunctationem meam regere. ep.
118 = 119 ὃ 3 rogo igiiur ut dubitationem
meam regere, id est beneficia tua interpretart,
ipse digneris. δὶ -- δῦ ὃ ὃ te, domine, rogo ut
me in hoc praecipue genere cognitions regere
digneris. 56 =64$ 1 summas, domine, gratias
ago quod inter maximas occupationes in ts
de quibus te consului me quoque regere dig-
natus es. 19=380§1 rogo, domine, consilio
me regas haesitantem.
quid et quatenus. 92 = 93 ut tu, domine,
dispiceres quid et quatenus aut permittendwm
aut prohibendum putares. cfs ΠΟ ΓΞΞ e
§ 1 quod an celebrandum et quatenus putes,
rogo scribas.
§ 2 nec mediocriter haesitavi. 118=119
§ 3 hic quoque non mediocriter haereo ne
cuiusquam retro habeatur ratio.
nomen ipsum, si flagitiis careat, an flagitia
cohaerentia nomwini puniantur. Lustin dial.
39. Tert. scorp. esp. 9. Iren. I 24 8 6 (of
Basilidians) quapropter et parati sunt ad
neyationem qui sunt tales, immo magis ne
122
pati quidem propter nomen possunt, cum
sint omnibus similes.
§ 3 perseverantes duci iussi. The late
Prof. H. J. Smith once asked me whether
capital punishment is of necessity implied
here. Doubtless, esp. after swpplicium. οἵ.
Sen. ep. 4 ὃ 9. de ira III 22 §2. Tert. ad
Scap. 5 Arrius Antoninus, when all the
Christians of a city presented themselves
before his tribunal, pawcis duci iussis reliquis
ait, & δειλοί, εἰ θέλετε ἀποθνήσκειν, κρημνοὺς ἢ
βρόχους ἔχετε.
pertinaciam Trajan ep. 57=65 ὃ 2 neque
enim sufficit ewm poenae suae restitur, quam
contumacia elusit. Gataker on Antonin.
ΧΙ ὃ 3 p. 386—7.
§ 4 quos quia cives Romani erant. adnotavi
in urbem remittendos. Trajan ep. 57 = 65
δ. 2 qui a Iulio Basso in perpetuum relegatus
est, . . . vinctus mitti ad praefectos praetorir
mei debet. 74 = 16 ὃ 2 quem ego perductum
ad me mittendum ad te putavi. T8=82 ὃ 3
si in urbem versus venturi erunt.
plures species inciderunt 56 =64 ὃ 4 nam
haec quoque species incidit in cognitionem
meam.
§ 5 imagini tuae 8=24 § 4. 9
WG ESE ek: “apocal,” 19. 10:
(Mommsen rom. Gesch. V 522).
ture ac vino supplicarent, praeterea male
dicerent Christo, Iustin dial. 131. Tert.
ady. Val. 30. Orig. in Matt. comm. ser. 38
(IV 267 Τὴ Basilidis quoque sermones detrah-
entes quidem tis, qui usque ad mortem certant
pro veritate, ut confiteantur coram hominibus
Iesum, indifferenter autem agere docentes ad
denegandum et ad sacrificandum dis alienis.
Epiphan. haer. 24 4. Haversaat’s criticism
that male dicerent belongs to ecclesiastical
Latin, would only hold if it were followed
by an accusative.
§ 7 secundum mandata tua hetaerias esse
vetueram 56=64 ὃ 3 mandatis tuls cautwm
est, ne restituam ab alio aut a me relegatos.
110-- 111 ὃ 1 wtebaturque mandatis tuis,
quibus eius modi donationes vetantur. 22 = 33
‘ mandatis mets. 90 -- 39 8 1 secundum man-
data mea.
§ 8 per tormenta quaerere. Torture of
slaves Eus. h. e. V 1 14 (ep. Vienn. et Lugd.).
Tren, fr. 13 (I 832 St.). Tustin apol. II 12.
Tert. ad nat. I 7.
dilata cognitione ad consulendum te decu-
eurri 72=77 rem integram distuli dum tu,
domine, praeceperis, quid observare me velis.
POS. TO SG 110s 2:
consulendum 56=64 ὃ 2 necessarium cre-
didi rem totam ad te referre. ὃ 5 per quod
effectum est, ut te consulerem. 29=38 § 1
quorum ego supplicium distuli ut te condit-
=25. 74
6 9.
THE CLASSICAL REVIEW.
orem disciplinae militaris firmatoremque con-
sulerem de modo poenae. 31=40§ 4 neces-
sario ergo rem totam, dum te consulerem, 77
suspenso reliqui. 58=66 ὃ 4 nihil decernen-
dum putavi, donec te consulerem de eo, quod
mihi constitutione tua dignum videbatur.
65=71 ὃ 2 ego auditis constitutionibus prin-
cipum, quia nihil inveniebam aut proprium
aut universale quod ad Bithynos referretur,
consulendum te existimavt, quid observari
velles. cf. 7T9=83 § 5. 68=73 te, domine,
maximum poniificem consulendum putavt.
81=85 ὃ 5 ego cum dandam dilationem et
consulendum te existimarem in re ad exem-
plum pertinenti. cf.82=86§1. 110=111
§ 2 quibus ex causis integram cognitionem
differendam existimavi, ut te, domine, con-
sulerem quid sequendum putares.
§ 10 victimarum, quarum adhuc raris-
simus emptor inveniebatur. Renan St. Paul
p. 308 n. 3. Philostr. Apoll. I 2. IV 41.
Schmidt Denkfreiheit 168. 181. Minuc.
12 $5.
ex quo facile est opinari quae turba homi-
num emendari possit, sisit paenitentiae locus.
That this last phrase may not be suspected
as biblical (Hebr. 12 17) cf. Liv. XXIV 26 §
15 ira deinde ex misericordia orta, quod adeo
Sestinatum ad supplicium neque locus paeni-
tendi aut regressus ab ira relictus esset. XLIV
10 ὃ 2 Andronicus . . traxerat tempus, id
ipsum, quod accidit, paenitentiae relinquens
locum. Fronto p. 207 Naber leviora sciens _
dissimulavit : locum paenitendi reliquit. dig.
XL 7 3 13 paenitentiae heredi locum non
esse. And Pliny himself I 24 ὃ 4 μέ paeni-
tentiae locum non relinguat.
ep. 97 (98) § 1 actum 27=88 pristinum
actum.
in excutiendis causis 86=82 § 2 ratio
totius operis excutiatur.
neque enim in universum aliquid quod
quasi certam formam habeat constitui potest.
113=114 honorarium decurionatus omnes qui
in quaque civitate Bithyniae decuriones fiunt
inferre debeant necne, in universum ὦ me
non potest statut. id ergo quod semper tutis-
simum est, sequendam cuiusque civitatis legem
puto. 65=71 § 2 nihil inveniebam aut pro-
prium aut universale quod ad Bithynos refer-
retur. 6672 ὃ 1 nec quicquam invenitur m
commentariis eorum principum qui ante me
fuerunt, quod ad omnes provincias sit con-
stitutum.
§ 2 suspectus in praeteritum. cf. 111=112.
115 = 116 mihi hoc temperamentum eius
placuit, ut ex praeterito nihil novaremus.
VIII 14 § 1 non ut in praeteritum (serwm
enim), verum ut in futurum, st quid simile
acciderit, erudiar.
THE CLASSICAL REVIEW. 123
sine auctore propositi libelli 96 § 5 pr.
31 = 40 ὃ 5 credibile erat neminem hoc ausum
sine auctore. 32 = 41 $1 qui damnati ad
poenam erant, non modo ea sine auctore, ut
scribis, liberati sunt.
nec nostri saeculi est 23 = 34 § 2 quod
alioqui et dignitas civitatis et saeculi tui nitor
postulat. 37 = 46 § 3 ego tllud unum affirmo
et utilitatem operis et pulchritudinem saeculo
tuo esse dignissimam. δῦ - 63 invitos ad
accipiendum compellere . . . non est ex iustitia
nostrorum temporum. δῶ = 86$1 cum pro-
positum meum optime nosses non ex metu nec
terrore hominum aut criminibus maiestatis
reverentiam nomini meo adquirere. 1 § 2
prospera omnia, id est digna saeculo tuo.
12 =7 felicitas temporum. 3 A = 20 § 2
tranquillitatt saeculi tui. cf. Τὰν. 4 68n.
M. Aurelius (Vuleat. Gall. Cass. 2) non
nostri temporis. Keim Rom p. 326. dig.
XLVIII 22 1. Dio LX VIII 6 8 4 (of Trajan)
διαβολαῖς ἥκιστα ἐπίστευε.
Mr. Hardy follows Schiller in rejecting
the evidence of ‘Tacitus and Suetonius
respecting the persecution of Christians by
Nero. Their positive testimony is supposed
to be outweighed by the silence of Seneca
and the elder Pliny. It is assumed that, if
either of these latter authors had heard of
Christians, they must have mentioned them
in their extant works. It would be charit-
able to assume that Dr. Schiller had read
neither Seneca nor Pliny. But he pushes to
an absurdity the argument from silence to
ignorance when he informs us (Nero p. 584
n. 3) ‘Juvenal also knows nothing of the
Christians.’ Marvellous as is the art with
which the poet of Aquinum packs into a
few pages a panorama of his age, guidquid
agunt homines, he necessarily leaves out more
than he can insert. The silence of Josephus
is no doubt significant, but the criticism
which infers from it ignorance of Christianity
certainly sins on the side of credulity rather
than of scepticism. Mr. Hardy goes farther
than Schiller. Not only does he reiterate
(pp. 51, 54, 55, 239-243) the assumption
that ‘Tacitus and Suetonius in relating the
events of Nero’s reign are speaking of the
Christians from the point of view of their
own time,’ which means, I suppose, that,
finding in their authorities ‘Jews,’ they
arbitrarily substituted ‘Christians’; but he
surmises (p. 211) that Pliny, when he says
(ep. 96 § 1) cognitionibus de Christianis inter-
Sui numquam, is guilty of the same confusion
as his two friends.
In Bithynia the Christians were especially numer-
ous, and Pliny on the spot would soon distinguish
them from the Jews, and applying this acquired
knowledge back to past events in which he then took
no interest, he describes as ‘ cognitiones christian-
orum’ (sic), what had been in the eyes cf the govern-
ment only trials of Judaisers.
Mr. Hardy seems inclined to believe that
Tacitus and Suetonius may have heard of
the Christians first from their common (or,
as he prefers to say, mutual) friend, Pliny.
P. 240. Dr. Lightfoot lays even more stress on
the language of Tacitus and Suetonius, who both of
them distinctly mention the Christians as victims of
Nero’s cruelty. The view that they may be ‘ inject-
ing into the incidents of the reign of Nero the
language and experience that belong to the age of
Trajan,’ he regards as a wholly gratuitous assump-
tion. That it is an assumption which cannot be
proved is not denied, but at least it has the advan-
tage of explaining the facts, which Dr. Lightfoot’s
assumption about the all-predominant influence of
Poppaea and her use of it against the Christians does
not do. Nor can it be said to be gratuitous. These
notices, written certainly not before the second decade
of the second century, are the very first indications
in non-Christian writers that the Christians were re-
garded as an independent body, or were in fact known
by name to the Roman world at all. Josephus,
Seneca, and the elder Pliny are absolutely silent
about them. This silence does not prove that those
writers knew nothing about the Christians, but it
certainly gives some grounds for the assumption,
while both Tacitus and Suetonius, whose notices about
the Christians are later in date than Pliny’s letter,
might have gained their information, if in no other
way, from their mutual friend.
An ingenious attempt, by aid of the ‘ great
might-have-been,’ to reduce three indepen-
dent witnesses to one. If Mr. Hardy will
look at the third chapter of the life of
Josephus, he will see that Dr. Lightfoot’s
statement respecting Poppaea is no gratuitous
assumption, like the rejection of the witness
of the two historians, or the argument from
silence, but rests on indisputable evidence.
The Jews, wishing to procure the release
of certain priests, send Josephus to Rome,
with instructions, as it appears, to approach
Poppaea by means of a favorite Jewish actor.
Josephus carries out his instructions with
perfect success. [Ὁ is incredible that then
for the first or last time Poppaea used her
influence on behalf of the Jews.
p- 28 (cf. p. 21) we read of Pliny: ‘A
dabbler in philosophy, he had no convictions
which could have drawn upon him the fate
of Thrasea, or Helvidius, or Rusticus.’ We
know that the informer Carus denounced
Pliny, and that Domitian’s death alone saved
him from sharing the fate of his friends
(ep. VII 24 § 14).
It is to be hoped that the following is not
an average specimen of Mr. Hardy’s use of
authorities (p. 215) :—
Ulpian lays it down, lib. viii, De officio Procons.
cap. 1, ‘servum alicuius corporis vel universitatis
124
torquere licebit in eorum etiam caput qui eius cor-
poris erant.’ Again, ‘Divus Hadrianus rescripsit a
suspectissimo incipiendum et a quo facillime potest
verum scire iudex crediderit.’
Few readers will take the pains to ascer-
tain that the latter quotation is from dig.
XLVII! 181 ὃ 2, where book eight of the
specified treatise of Ulpian is cited, but not
chapter one. The former quotation, though
given in inverted commas, contains, not the
precise text of Ulpian, but Baudouin’s sum-
mary of his teaching (Mr. Hardy himself
corrupts licebat to licebit). To Baudouin also
is due ‘cap. 1,’ for which no authority exists.
At least none is known to Otto Lenel
(Palingenesia iuris civilis, Lips. 1889, 11
966-991) who has collected with great exact-
ness the extant fragments of Ulpian’s book.
A somewhat similar freedom is taken with
Ulpian’s words on ep. 96 ὃ 1 p. 2118:
The general duty which, as Ulpian says, was in-
cumbent on all provincial governors, ‘statim atque
in aliqueim (516) celebrem civitatem vel provinciae
caput proconsul venit, debere sedes sacras circumire
atque inspicere.’
This is from book II of the de officio pro-
consulis (11 968 Lenel, dig. I 16 7 pr.) si in
aliam quam (v. 1. aliquam) celebrem civitatem
vel provinciae caput advenerit,... debet ...
THE CLASSICAL REVIEW.
§ 1 aedes sacras et opera publica circumire
inspiciendi gratia.
10 would be easy to add illustrations which
have escaped the commentators. Thus ep.
10 8 1 epistulae is used like litterae and
ἐπιστολαί, of a single letter. See my note on
1119 ὃ 13 and ind. ep. 23 ὃ 2 saeculi tui
nitor. Tac. d. 22 daetitiam nitoremque
nostrorum temporum.
In conclusion I would recommend the
letters regarding the Christians as a worthy
theme for a monograph. Any one who will
digest the whole literature will, I believe,
prove beyond possibility of cavil, that it
would have required a more skilful forger
to compose the correspondence of Pliny and
Trajan than to compose the nine books of
miscellaneous correspondence ; and that, as
Aldus said long ago, letters 96 and 97, far
from being an interpolation, establish the
genuineness of the entire collection.
[| P.S.—The best criticism of these letters,
known to me, is the latest—in Neumann’s
Der rimische Staat und die allgemeine Kirche
bis auf Diocletian (Leipz. Veit), of which
the first volume has just appeared. |
Joun E. B. Mayor.
BURY’S HISTORY OF THE LATER ROMAN EMPIRE.
A History of the Later Roman Empire, from
Arcadius to Irene (395 a.p, to 800 a.D.),
by J. B. Bury M.A., Fellow and Tutor
of Trinity College, Dublin. (Macmillan &
Co. : 2 vols. 8vo.)
Mr. Bury’s volumes are an important and
valuable contribution to our knowledge of a
period, the history of which has been too
much neglected by scholars. It may be
safely said that, for a student of the History
of the World, Imperial Rome—whether we
apply that name to the city by the Tiber or
to the daughter-city by the Bosphorus—is
far more important than Republican Rome ;
yet how many an undergraduate who knows
every detail of the Second Punic War would
fail to give a satisfactory account of the
great event which marked the reign of
Constantine Pogonatus !
Of course this partial treatment of history
is to some extent explained by the immense
difference between the literary excellence of
the earlier and the later historians. It is
far pleasanter to follow the footsteps of
Hannibal under the guidance of Livy or
Polybius than to piece out the history of
Heraclius from the turgid sentences of
Theophylact or the curious iambics of George
of Pisidia. But if history is to be studied
seriously as a science 1t will not be possibile
for the student to confine his reading to
books distinguished by their literary ex-
cellence ; and the more repulsive the material
the greater is the credit due to an author
who, like Mr. Bury, has patiently and labor-
iously dug it for us out of the mine and
wrought it into historical form.
It will be seen from the title that the book
deals with the four great transitional cen-
turies between Theodosius and him whom
we used without fear to speak of as Charle-
magne, that is to say between the so-called
‘Downfall of the Western Empire’ and its
restoration by the King of the Franks.
How much there was that was wrong and
misleading in this view of the matter is fully
explained by the author in his preface. He
is a loyal follower of Bryce and Freeman, and
it may be said that one great purpose of his
book is to show the thorough continuity of
the Empire, and to explain how completely
THE CLASSICAL REVIEW.
during these transition-centuries Augustus
Imperator remained the foremost figure in
the eyes of the European nations, whether
he held his court on the Palatine at Rome or
by the Augusteum at Constantinople.
It would be interesting but is of course
here impossible to discuss the author’s views
of some of the chief actors during these
eventful histories. His bias is throughout
rather more pro-Roman and anti-Teutonic
than that of some recent authors, including
the writer of this notice. Thus he takes
the least favourable view of Stilicho’s char-
acter, considering him little better than an
accomplice with Alaric in his invasion of
Italy and believing the story told by the
ecclesiastical historians of his ambitious
views on behalf of Eucherius his son. So
too he speaks of ‘ the tender hot-bed plant of
Theodoric’s Ostrogothie civilitas which had
never really looked promising’: a remark
from which some admirers of Theodoric
would venture to dissent, holding that
Theodoric’s system was full of promise for
Italy and the world, and that, if it had had
anything like fair-play, the barbarian dark-
ness of the Middle Ages might have been
shortened by some centuries. He also has
little but praise for Justinian, the great re-
asserter of the supremacy of ‘ Romania’ as
against ‘ Barbaricum,’ and entirely dissents
from the view that his conquest of the
Vandal kingdom in Africa smoothed the
way for the victories of the Saracens in
that region. All these paragraphs would
furnish material for much interesting con-
troversy if this were the time or the place for
it. There is certainly much force in Mr.
Bury’s argument that the fact that Carthage
formed part of the Empire at the critical
moment of the dethronement of Phocas was
of vital importance to the safety of the
Empire. From this point of view the best
champion of the fame of Justinian is
Heraclius.
For readers of this Review probably the
most interesting chapters in the book will
be those which relate to the history of
literature and the survival of the Hellenic
nationality. Mr. Bury has devoted great
pains to this part of his subject and we
think has treated it very successfully. His
remarks on Procopius are very just.
‘In fact Procopius was at core, in the
essence of his spirit, a pagan: Christianity,
assented to by his lips and his understanding,
was alien to his soul, like a half-known
foreign language. He could not think in
Christian terms: he was not able to handle
the new religious conceptions, he probably
125
felt wonder, rather than satisfaction, at the
joys that come from Nazareth. And we
may safely say that it was just this pagan
nature, deeper perhaps than that of the
aggressive Zosimus, that (in the peculiar
circumstances of the times) made him such
a good historian. He is almost worthy to
be placed beside Ammianus.’
We are rather surprised to read that
‘there is no reason to consider the “ Edifices ”
of Procopius an insincere work, although it
was perhaps written to order.’ We should
have thought it, as the work of the same
pen that wrote the ‘ Histories ’—to say no-
thing of the ‘ Anecdota’—one of the most
thoroughly insincere books in literature.
On the difficult question of the authorship
of the ‘ Anecdota,’ Mr. Bury is disposed to
follow Ranke rather than Dahn, and to
attribute this strange performance to a sort
of clique of opposition scribes who may per-
haps have got hold of some ‘ MS. of frag-
mentary jottings written by the true Pro-
copius’ and tacked on to these their wild
and scurrilous stories against the Emperor
and his court. The suggestion is an ingeni-
ous one, but it will require a good deal of
argument to overthrow the strong case for
the absolute identity of authorship made
out by Dahn in his ‘ Procopius von Caesarea.’
Possibly the ultimate conclusion will be as
to this most difficult question—
‘All that we knowis, nothingcan be known—’
but we should like to suggest yet one
more possible theory to the disputants, and
that is that the ‘Anecdota’ may have
been the work of Procopius under the influ-
ence of a disordered intellect.
Mr. Bury has much to say on the obscure
but important subject of Slavonic settle-
ments in Greece. He accepts in the main
the argument of Hopf and rejects the famous
theory of Fallmerayer which would have
dethroned the present inhabitants of Greece
from their Hellenic pride of birth and made
them descendants of Slavonic barbarians.
He points out however some flaws in the
reasoning of Hopf, whom he considers
‘almost as much an advocate as Fallmerayer,’
and concludes that, at any rate in the eighth
century, there was a large infusion of the
Slavonic element, rather perhaps in the
country districts than in the towns of Greece,
and that at any rate the earlier Slavonic
settlers very speedily adopted the Hellenic
speech and the Christian religion.
For the remarks of the author as to the
changes which the Greek language under-
went we must refer our readers to his very
126
interesting chapter (Book 1V., cap. vii.) on
‘the Language of the Romaioi in the Sixth
Century.’ He distinguishes ‘ three kinds of
Greek : (1) the vulgar spoken language from
which modern Greek is derived, (2) the spoken
language of the educated, which under the
influence of the vulgar tongue tended to
degenerate, and (3) the conventional written
language which endeavoured to preserve the
traditions of Hellenistic prose from the
changes which affected the oral “ common
dialect.””’ He concludes, partly from the
strange dialogue between Justinian and the
Green Faction which has been preserved by
Theophanes, that the Hellenic language was
‘already far on its way to becoming what is
called Romaic and was in fact already known
by that name.’
‘A sixth-century inscription in Nubia
proves that the word νηρόν was then used for
“ water,’ whence comes the modern Greek
νερό. A mule is βορδώνης instead of ἡμίονος
and σγαυδάριν or yavddpw is apparently used
THE CLASSICAL REVIEW.
for an ass. A standard is Bavdov, an iron-
headed club is δίστριν, luggage is τοῦλδον,
and σκούλκα is used for a guard or watch.
Besides the strange vocabulary, derived
partly from Latin and partly from local
Greek words, changes are taking placein the
grammar and syntax. Terminations in -ἰον
for example are becoming corrupted to -w:
the perfect tense and many prepositions and
particles are falling into disuse.’
We conclude, as we began, by commending
to the attention of all historical students, but
especially of those who may have been
chiefly occupied hitherto with the fortunes
of Athens and the Elder Rome, this careful
and patient survey of the history of the
Roman Empire during a _ period which
witnessed changes of the most momentous im-
port to the nations of Europe and Asia, the
effects of which we are continually feeling in
the political controversies of our own day.
Tuos. HopcGKInN.
RAWLINSON’S HISTORY OF PHOENICIA.
History of Phoenicia. By Grorcr Raw in-
son, M.A., Camden Professor of Ancient
History in the University of Oxford,
London and New York: Longmans, Green
and Co: 1989. Pp, xxii” 89. ὃὅνο.
24s.
ΤῊΝ Bible is the chief source of Prof.
Rawlinson’s information about Phoenicia :
he cites it more than 250 times as his
authority. Its evidence appears to him
conclusive, and is accordingly exempted from
any criticism. Thus he says on p. 449 :—
‘The army which he [Sennacherib] set in
motion must have numbered more than
200,000 men.’ And in a note he gives his
reason for this estimate :—‘ It was the same
army which lost 185,000 men by miracle in
one night (2 Kings, xix. 35) And yet he
at times fails to take note of important
biblical statements ; e.g. in speaking of the
siege of Tyre by Nebuchadnezzar, he says on
Ρ. 473 that the city submitted :—‘ Thus
Tyre, in B.c. 585, fell from her high estate.
Ezekiel’s prophecies were fulfilled.’ But
Ezekiel himself, xxix. 18, admits the notori-
ous fact that his prophecies were not fulfilled,
inasmuch as the siege failed.
The Egyptian and Assyrian sources have
not been properly utilized. For Egyptian
sources, there is one reference to Brugsch’s
History of Egypt, and three references to two
translations in Lecords of the Past: and that
is all. Consequently on pp. 406, 407, there
is only some vague talk about Egyptian
supremacy and its decline, where there -
should be precise statements about the
Egyptian invasion of Phoenicia in the
twenty-ninth year of Thothmes III., the
invasion of Egypt from Phoenicia in the
eighth year of Rameses ILI., and so forth.
There is no allusion to the alleged Phoeni-
cian origin of the Hyksos ; nor to the sug-
gested identification of the Fenchu of the
inscriptions with the Phoenicians. For
Assyrian sources, there are only references
to George Smith’s Hponym Canon, and to
translations in Records of the Past. Con-
sequently many historical events are com-
pletely ignored : e.g. the reduction of Cyprus
by Sargon, attested by the Larnaka inscrip-
tion. The Babylonian or Assyrian origin
of Phoenician writing, maintained by some
high authorities, is likewise ignored ; though
space is found on p. 378 for wild conjectures
about Hittite hieroglyphic and the Cypriote
syllabary. Curiously E. de Rougé’s theory of
derivation from Egyptian hieraticisattributed
by Prof. Rawlinson cn p. 377 to a writer in
the last edition of the Hncyclopaedia Britan-
nica,as if it were there enunciated for the first
time.
THE CLASSICAL REVIEW. 127
The Greek and Latin sources were pretty
completely collected by Movers; and then
Kenrick appropriated most of his results.
But Prof. Rawlinson has been imprudent in
trusting entirely to Kenrick. For instance,
Movers made a perfectly convincing emenda-
tion of sixteen for sixty in Josephus, Anti-
quitates, ix. 14. 2, but Kenrick missed this ;
and accordingly Prof. Rawlinson has made
mistaken observations on pp. 446, 447, in
the belief that the number really was sixty.
Such of the Greek and Latin sources as have
been utilized, have not been properly sifted :
the trustworthy and untrustworthy being
cited with provoking impartiality. And
many of them have not been utilized at all.
There is no allusion, for example, to the
so-called Treaty of Kallias: though the pros-
perity and foreign commerce of Phoenicia
must have been profoundly affected by an
agreement which secured the Aegean to the
Greeks and the eastern waters round Cyprus
to the Persians and their dependents the
Phoenicians.
The material sources of information about
Phoenicia have lately been discussed by
MM. Perrot and Chipiez in the third volume
of their work: and their results might have
been subjected to judicious condensation.
As it is, some things are noticed at length,
while others of equal importance are dis-
missed in a few words or simply ignored.
Thus, exactly one tenth of the whole space
devoted to architecture and art and manu-
factures is occupied by a detailed description
of a series of bowls: and this is mere de-
scription in the style of an auction catalogue,
quite valueless apart from criticism. The
illustrations, with hardly an exception, are
taken from this volume of MM. Perrot and
Chipiez. The obligation is generally ac-
knowledged : but on p. ix. of the preface and
in the list of illustrations it is stated that
many of them are taken from various other
books there mentioned. This is misleading :
they were taken from those books by MM.
Perrot and Chipiez. There are similarly
misleading statements on pp. vii. and ix. to
the effect that Prof. Rawlinson has drawn
materials from Movers. He has them at
second hand through Kenrick.
For Phoenician inscriptions Prof. Rawlin-
son has relied, in every case but one, upon
the translations in Gesenius and in the early
parts of the Corpus Inscriptionum Semitic-
arum. But hes quotes the inscription of
Tabnith on p. 395 from his own book, 716
Story of Phoenicia. Had he referred to his
real authority, M. Renan’s translation in the
Revue Archéologique for July 1887, he would
have seen that he has left out a line.
On pp. 396, 397, he gives a rendering of
C.ILS. No. 1: and says in a note that he
has followed M. Renan’s translation. Here
is a portion of this translation :—‘ Omnis
regia stirps et omnis homo qui addiderit
facere opus super al(tare hoc et super]
coelaturam auream hance et super porticum
hance, [pono] ego Iehaumelecus, [rex Gebal,
faciem meam adversus eum qui] fecerit opus
hoe, et sive...sive...istum., Et quicunque...
super locum hune, [et quicunque..., aboleat]
domina Baalat-Gebal hominem ipsum et
semen ejus.’ And here is the rendering of
this :—‘ Every royal personage and every
other man who shall make additions to this
altar, or to this golden carving, or to this
portico, I, Jehavmelek, king of Gebal, set
my face against him who shall do so, and I
pray my lady Beltis of Gebal to destroy that
man, whoever he be, and his seed after him.’
The rendering, whoever he be, is really
admirable.
The slipshod reasoning on imperfect or
inaccurate information, which is too charac-
teristic of this book, cannot be fully exposed
in a brief review. The opening statements
of the book must serve here as specimens
of its style throughout. Prof. Rawlinson
begins by saying that the palm-tree ‘ formed
a leading and striking characteristic’ of
Phoenicia: and hence the Greeks ‘called
the tract Phoenicia, or the Land of Palms;
and the people who inhabited it the Phoeni-
cians, or the Palm-tree people.’ Now, it is
true that Φοῖνιξ means a Phoenician and also
means a palm-tree, and that Φοινίκη means
Phoenicia. But the argument from mere
identity of name goes just as well the other
way :—the Greeks called the palm-tree the
Phoenician because it was characteristic of
Phoenicia. And then the meaning of the
name must be considered. The Phoenicians
may have got the name because they were
themselves of ruddy countenance, or because
they traded in the ruddy dye; but palm-
trees are not strikingly ruddy. He then
argues that the palm must anciently have
been among the chief ornaments of the coast
round Aradus, becanse ‘the palm is the
numismatic emblem of Aradus.’ It is only
on one small series of the coins of Aradus,
from 244 to 183 B.c., that the palm occurs
as anemblem. But the inference would be
unwarrantable in any case, for such emblems
were chosen quite capriciously.
κοι, Torr.
128
SCHMIDT'S
HANDBUCH DER LATEINISCHEN
THE CLASSICAL REVIEW.
UND GRIECHISCHEN
SYNONYMIK.
Handbuch der Lateinischen und Griechischen
Synonymtk, von Prof. Dr. J. H. Hetnricu
Scumipt. Leipzig: B. G. Teubner. pp.
844, 12 Mk.
In the Preface to the fourth volume of his
Synonymik der Griechischen Sprache (four vols.
1876-1886) Prof. Schmidt promised a ‘ hand-
book’ of synonymes, which now appears in
the form of this stout octavo. The reason for
the delay in its appearance is somewhat
amusing : the author has been plunging deep
intonatural science with the help of a‘splendid
microscope, and satisfying himself of the
worthlessness of materialistic theories. He
offers us some comments upon the usage of
some 2400 Greek words, and about 1500
Latin words, arranged according to their
meaning in 126 groups. But he does not
deal with these by definitions or ‘ Schlag-
worter’; this would be unworthy of an
honourable man, most definitions being in-
complete and often untrue. He prefers to
give, at greater length than some of his
predecessors, illustrative quotations, not
however sparing to give his own conceptions
of their force. In Greek Dr. Schmidt has
not merely abridged his larger work ; he
has added to it in some respects, though of
course omitting much. In Latin he has
been more dependent upon others, especially
on Doderlein, from whom however he often
differs. His own original contributions have
been mainly drawn from natural science, but
here he has been concerned rather to stim-
ulate the reader than to exhaust the subject.
Certainly Dr. Schmidt’s method gives much
freshness to his style ; and he is justly proud
of having now published some 7000 pages
without a footnote !
A work of this magnitude can hardly be
reviewed satisfactorily, except after it has
been in frequent use for months or years ;
but a few points of interest may be noted,
which have presented themselves on a
cursory inspection.
Distinguishing between flavus and fulvus,
he finds the former to be a yellow, the latter
a brown. But when we have the ‘brown
beak’ of the nightingale (Ar. Av. 214 etc.)
he bids us remember that the Greeks were
fond of transferring words of colour to
motion, and he would have us translate
rather ‘ quivering,’ just as Babrius applies
ξουθός to the ‘steel-blue’ swallow, and
Chaeremon even to the wind. Fulvus he
would allow to include the whole sphere of
flavus (except pale yellow), and brown as
well. Zuteus is defined as a yellow passing
into red, an orange red; and we are told
that the strict meaning is determined by the
fact that the yolk of an egg is called Jutwm
(more correctly /utewm); while croceus is a
pure yellow not inclining to red. But what
are we to make in that case of Virgil’s
eroceo mutabit vellera luto? If Ovid applies
luteus to sulphur, this is put-down as an
exaggeration: and Horace’s luteus pallor
is pronounced ‘an intentional jest.’ Surely
this is to attempt a precise limitation
which the usage of the language does not
permit. Elsewhere ferrugo is simply defined
as ‘steel-blue,’ without any attempt to dis-
cuss the various uses of this most perplexing
word, For χλωρός Dr. Schmidt will allow
no literal meaning but ‘green’: when used
of honey and cheese, he insists on the
rendering ‘fresh,’ and the χλωρὰ ψάμαθος of
Soph 47. 1064 is for him the ‘ grass-grown’
sand. Few will agree with him that a
poet could not have brought in such an
‘ugly’ picture as a corpse on ‘ yellow’
sand. Here again there is an attempt at
quite needless precision. The treatment of
unda (κῦμα) and fluctus (κλύδων) is much
more satisfactory than Doederlein’s. Dr.
Schmidt’s criticism of Doederlein’s comments”
on emort is convincing ; evoev and ὑπνοῦν are
also well treated. He is less successful in
his attack upon the accepted distinction
between potentia and potestas; none of his
examples debar us from confining the latter
to lawfully conferred or at least natural
authority, which of course may be unlimited
in its character. The difference between
incipere and incohare is brought out by
illustrations which are none the less easy to
remember because they are rather comical.
On the whole the derivations are in accord-
ance with modern science, herein contrasting
very favourably with some current text-
books ; but it may well be doubted whether
χορός is originally the dance in which all who
took part in it clasped (root XEP) each other’s
hands, and not rather the enclosed spot for
dancing (cp. Homer’s λείηναν δὲ χορόν). It
would not be difficult to find other points
of difference or of doubt ; but the book is
both a useful and an interesting one, and
deserves to be welcomed accordingly.
A. S. WILKINS.
THE CLASSICAL REVIEW.
Homer’s Odyssey. Books I.—IV. Edited on
the basis of the Ameis-Hentze edition, by B.
PERRIN, Professor in Adelbert College of Western
Reserve University. Boston, U.S.A. Published
by Ginn & Company, 1889. [College Series of
Greek Authors edited under the supervision of
JoHN WILLIAMS WHITE and THomAs D. Sry-
MOUR. ]
KARL FRIEDRICH AMEIS was no ordinary man. He
laboured all his days in the Gymnasium, and his
commentary on Homer is the only important work
by which he is remembered. Of this commentary he
only lived to complete the Odyssey and the early
books of the Iliad. But his commentary differs from
most other German commentaries in the degree to
which the author’s personal interest and personality
reveal themselves in his work. Ameis had abundant
erudition and he availed himself of all sources of
knowledge within his reach, not least important
among which was the co-operation of a large number
of personal friends. He knew and loved his Homer,
he makes Homer his own interpreter, and his loving
acquaintance with the great poems yields to the
readers of his notes ever fresh pleasure and fresh
knowledge.
Ameis’s edition, incomplete as regarded the [iad
at his death, was carried forward by Dr. Hentze of
the Gottingen Gymnasium, an exact and learned
scholar, and the Ameis-Hentze edition has been made
the basis of this volume of the College Series of Greek
Classics which it is my privilege to review.
This American book is the fruit of an unusual
union of effort by different persons, each of the
highest competence, and Professor Perrin wiil be glad
that attention should be called to the fact that the
preparation of a portion of the commentary (on a
and a part of 8) had been made by Professor Packard
before his recent lamented death. It would not be
easy to name a book, carried over from the German
into English, in which so few traces of its German
origin appear. This results not only from the
abundant amount of fresh material which Profes-
sor Perrin has introduced, but from the skill and
taste characteristic alike of Professors Perrin and
Packard.
The best Homeric scholars will find the most
pleasure in perusing this commentary. This volume
has been prepared with reference to use in schools as
well as in colleges. Few Greek classics are better
adapted for use in the higher forms of classical schools
than the Odyssey. It is an unequalled mirror of
Greek life. Its narrative form attracts and allures
the young scholar, while its simplicity of style places
no difficulty in his path. Professor Merry’s admirable
editions of the Odyssey have richly deserved the
success they have gained, but the present volume will
be found to supply material not contained in Merry’s
commentary.
Rosert P. ΚΕΒΕΡ,
129
Das Praesens der indogermanischen Grund-
sprache, von Orro HorrMANN. Gottingen, Van-
denhoeck und Ruprecht : 8. 145.
Dr. HorrMANN’s treatise 15 ἃ useful reminder that not
everything written in German is to be implicitly fol-
lowed by the trustful philologist, even though its
preface bear the date ‘mai 1889’ and its formulae
bristle with yand τῳ. If we may trust our present
evidence, the author’s principles were formed under
the old school, though stress of circumstances compels
him to use some modern terminology and to quote
Brugmann with respect. But this sop to Cerberus is
sufficient, and a notation conformed to Fick’s WV dérter-
buch hints to the initiated that the mystic symbols
of more recent invention must not be taken too
seriously. Indeed a doubt has now and then insinu-
ated itself whether certain dangerous neologians have
in all cases even been read, but then are we not told
in the preface that references and polemic are neces-
sarily excluded in a compendium ? It is hardly need-
ful to go into detail over so disappointing a perform-
ance, but a few criticisms may be offered to save us
from dismissing a bookful of ipse dixits in one cate-
gorical statement of the same kind.
We may begin with some samples of the results
which our author has, no doubt after mature reflex-
ion, ignored. Brugmann’s paper in the JJorpholo-
gische Untersuchungen on the Sanskrit verbs of the
yundjmi type has not prevented ‘ju-né-g-mi’ from
appearing asa genuine I.E. form ; nay, more, we have
a new theory of the other non-thematic nasal classes
built on this eminently sound foundation, which only
needs the reversal of the principles of vowel-contrac-
tion to make a very pretty house of cards indeed.
The ‘ aorist-present az6’ (sic) keeps its place, but we
are not told what is the weakness of Hiitbschmann’s
argument against it. The treatment of ‘ /és (sit)’
shows no trace of borrowing from Osthoff’s Perfekt.
Moreover, without being very exacting, we might
have expected some reference to sundry important
papers which came out in good time for use, especi-
ally Wackernagel on the first aorist passive (‘ ἐδόθης Ξε
ddithas), and Zimmer on the Italo-Keltic ‘passive’
7. A few sins of commission may be added to pair off
with these specimens of omission. The sonants may
become anything they please: 7 is Gk. :pand pv, Lat.
ri and rd, while ἢ) may be recognized in Gk. as ιν.
Metathesis is presumably responsible for the equation
between Skt. deyati and ὀπυίω. Prothesis is very
freely used. In ‘a-vw, α-ἴδομαι,᾽ etc., it is not true,
but neither is it new; in ‘Skt. dredmi =*redmni
mit prothetischem a’ we may fairly congratulate Dr.
Hoffmann on a new idea. Contraction brings sim
out of siem, Vedic jigat out of *ji-gi-nt, and δάμναμι
out of ἔδαμ-νέ-α-μι. The analogy of χθὲς and ἱκτῖνος
proves that -diydi would become -θθαι, -σθαι: μέσος
is an analogy not allowed to count. We may close
with a real discovery, adumbrated in the cipher
“ «ldvarar=Kd-v-a-ra’ (p. 136). A vocalic 5! Here is
indeed a treasure. Will Mr. Hoffmann send us a
phonogram to indicate its pronunciation 4
J. H. Mounron.
NOTES.
Nore ON THUCYDIDES, v. 68 jin.—ém) δὲ βάθος
ἐτάξαντο μὲν ov πάντες ὁμοίως, ἀλλ᾽ ὡς λοχαγὺς ἕκαστος
ἐβούλετο, ἐπὶ πᾶν δὲ κατέστησαν ἐπὶ ὀκτώ. This
Statement is puzzling, and has been discussed at
length by Grote and others. It may mean, as Pro-
NO. XXXI. VOL. IV.
fessor Jowett suggests, that in some cases one part of
the line was deepened at the expense of another, and
the rear rank of one ἐνωμοτία posted behind another.
Or possibly the ranks behind the first did not in-
variably contain four men. Thus an arrangement of
K
130
4, 3, 4 etc. would give 9 rows amounting to 32 men
in all. But we must remember that Thucydides is
not so much discussing the order of the Spartan sol-
diers, as furnishing an estimate of their numbers.
He ‘could not do this exactly,’ but he guarantees
three things, the number of ἐνωμοτίαι, viz. 112;
the number of men, 448, in the front rank; and
the average (ἐπὶ πᾶν) depth, viz. 8. Now, if the
number of men in each ἐνωμοτία was the same, we
have not merely an approximate calculation, but an
exact one, the very thing which Thucydides dis-
claims. May it not then have been the case that, in
one or more of the λόχοι, the ἐνωμοτία consisted of a
number more or less than the standard 32? This
would be a part of the ‘system of secrecy’ (ris
πολιτείας τὸ κρυπτόν) ; and if it was privately man-
aged by the military authorities we can see at once
how the total number was ‘kept dark’ (ἠγνοεῖτο).
According to the text this arrangement was left to
the respective Aoxayot: but it may be that the clause
ἀλλ᾽ ὡς λοχαγὸς ἕκαστος ἐβούλετο is spurious, as
Dobree suspected. The ἐνωμοτία or ‘ section,’ which
was the unit of the Spartan army, undoubtedly
varied in size at different periods ; and may possibly
have varied in different λόχοι or ‘regiments’ at the
same period. This explanation seems not unreason-
able, for if Thucydides had been sure of the number
of men in each ἐνωμοτία, why did he not state it, in-
stead of only giving the number in the front rank ?
C. E. GRAVES.
* *
*
CATULLIANUM.—Altis 62, 3.
Quod enim genus figuraest ego non quod obierim ?
Ego mulier, ego adolescens, ego ephebus, ego puer,
Ego gymnasi fui flos, ego eram decus olei.
Many proposals have been made to emend mu/ier,
as ego puber, Scaliger; ego iwuenis, Rosekerg ; ego
enim wir, Postgate ; but none which 1 have seen seems
to me probable. I think that either mzutlier is sound,
or miles should be substituted for it. O has mulieés,
but too much weight is not to be attached to this.
Miles would denote the fact that Attis had just arrived
at military age, that of early manhood, not necessarily
older than eighteen.
A. PALMER.
x %
*
Propertivs II, xxv. 19, 20.
ultro contemptus rogat et peccasse fatetur
laesus et inuitis ipse redit pedibus.
The editors find no difficulty in inuitis...pedibus,
but the words seem to me to strike a false note. Of
course, expressions of this kind are suitable enough
when they are used of a man halting between two
opinions or of an inward desire repressed by a strong
mental effort. Thus, for instance, in Ov. Rem. Am.
218 sed quanto minus ire uoles magis ire memento:
perfer et inuitos currere coge pedes we have described
the lover’s resolution to tear himself from the em-
brace of his mistress. But there is no room for a
double purpose here: on the contrary, the coldness
and fickleness of Cynthia are contrasted with the un-
wavering fidelity of Propertius. Whatever the exact
meaning Οὗ]. 17, it is clear that the poet's love is
declared to be less susceptible of change than iron or
granite under the influence of rust or dripping water.
Where then is the justification of inuitis pedibus ?
By substituting inuisis for inuitis, we obtain a perfect
parallelism between ultro rogat, peccasse fatetur, and
ipse redit on the one hand and contemptus, laesus,
and inuisis pedibus on the other. It should be added
THE CLASSICAL REVIEW.
that F has inuictis, but the variation is unimportant :
cf. 1. ii. 1 uita, F uicta; I. vii. 18 aeternos, F ecternos
and many others.
Propertius II. xx. 25.
ergo qui pactas in foedera ruperit aras.
So all Baehrens’ MSS. although G has actas and P
fractas. Hertzberg defends pactas aras in foedera
‘altar pledges given in troth’ by Ov. Met. ix. 722,
but, if the context be examined, it will be found that
pactae taedae means simply ‘the marriage which had
been agreed upon.’ On the other hand, tangere aram
is the regular phrase in these circumstances: Juy.
xiii, 89 quaecunque altaria tangunt ; xiv. 219 Cereris
tangens aramque pedemque; Liv. xxi. 1 Hanni-
balem...altaribus admotum tactis sacris; Verg. Aen.
xii. 201 tango aras. But for the harshness of
ruperit aras, it would be easy to restore tactas...aras
with most editions before Lachmann, who, while
objecting to the use of rumpere, found less difficulty
in the combination pactas aras rumpere. But it must
first be shown that pactas aras is possible Latin.
Perhaps we should read :—ergo qui tacta sua foedera
ruperit ara: I find that tacta sic...ara has already
been suggested by Mr. Housman. At the same time,
it is difficult to affirm that tactas aras rumpere is
impossible for Propertius: hardly less bold is LV. xi.
3 cum semel infernas intrarunt funera leges. Simi-
larly, in place of the normal παραβαίνειν νόμον, ὅρικον
«.7.A., Herodotus has (vi, 12) τίνα δαιμόνων παραβάντες
τάδε ἀναπίμπλαμεν ;
A. C. PEARSON.
Med
On Mr. SEATON’s ‘ITERATIVE USE OF ἄν, Class.
Rev., October, ’*89.—Omitting the reconciliation of the
unreal or contrary-to-fact notion with the ideal or
potential notion, J offer two remarks :
(1) The term ‘condition’ asserts the simple rela-
tion of antecedence and consequence—the latter being
necessary. Whether this relation is always, some-
times, never, fulfilled, or whether the fulfilment is
uncertain—these practical accidents have nothing to
do with the theory of condition.
Moreover, the general—time, place, or term—is
essentially conditional of its special applications : so
the general statement, actually true as such, ‘A
drowning man catches at straws,’ is conditionally
true of any special ‘ drowning man.’ Hence to substi-
tute ‘indefinite temporal’ for ‘conditional’ in itera-
tive sentences is only to take the species for the
genus, and does not get rid. of the conditional ac-
knowledgment.
(2) There is no difference between the phenomena
reported by the imperfect aud the aorist, except ‘in
the mind of the narrator.’ The aorist gives the
simple yea or nay of occurrence, without taking
account of the historical or spectacular incidents ;
but it does not deny the existence of these in ignor-
ing them to get at the upshot. Hence, the aorist
carries latent what the imperfect exposes : so that the
use of both for iteration isnot remarkable. Of course,
the presence of ἄν only creates or emphasizes explicit-
ness out of the implicit, since the gnomic aorist and
the ordinary imperfect already involve iteration with-
out ἄν. This explanation is distinctly opposed to
Goodwin’s (Gk. Gr., ὃ 205, 2, N. 1), which is indeed
self-contradictory. ‘One distinct case’ consorts as
badly with πολλάκις as with οὔπω, as iteration for or
against is inevitable. No doubt, the explanation is
to be sought in Goodwin’s conception of ‘ animated
language.’
CASKIE HARRISON.
THE CLASSICAL REVIEW.
‘To ταν WorpswortH. Composed
for the greater part on the same night after
the finishing of his recitation of the poem
in thirteen books, on the growth of his own
mind’ (vide Knight’s Life of Wordsworth, 11.
p. 184).
: : : thy faithful hopes,
Thy hopes of me, dear friend! by me unfelt !
Were troublous to me almost as a voice,
Familiar once and more than musical :
As a dear woman’s voice to one cast forth,
A wanderer with a worn-out heart forlorn,
’*Mid strangers pining with untended wounds.
O friend! too well thou know’st of what sad
years
The long suppression had benumbed my soul,
That, even as life returns upon the drowr’d,
The usual joy awoke a throng of pain—
Keen pangs of Love, awakening, as a babe,
Turbulent with an outcry in the heart !
And fears self-willed, that shunn’d the eye
of hope,
And hope that scarce would know itself from
fear ;
Sense of past youth and manhood come in
vain,
And genius given and knowledge won in
vain ;
And all, which I had culled in wood-walks
wild,
And all, which patient toil had reared,
and all
Commune with THEE had open’d out—but
flowers ;
Strew’d on my corse, and borne upon my
bier,
In the same coffin, for the self-same grave.
. . .
S. T. CoLERIDGE.
131
κεκαρτέρηται Tapa.
Ὁ > A
Ais δ᾽ ἐλπίσιν σὺ τοῦδ᾽ ἀνελπίστου, φίλε,
πιστὴν ἐβόσκου καρδίαν καρπούμενος,
CART) .9} td ’, ΡΣ
aid ἠρέθιζόν μ᾽, ὥσπερ εἰ φθόγγον ποτὲ,
+ 3, ee / id ’ ,
εὔγνωστον ὄντα πρόσθεν, ἡδίω λύρας,
΄ \ >
φθόγγον γυναικὸς εὖ φιλουμένης, φυγὰς
΄ see rn
κλύοι τις ἀξένοισιν ἐνσεισθεὶς ὁδοῖς,
ζι
A 5 \ vo
ὅς ἀμβλὺς ἤδη τἄνδον, οἰόφρων πλάνης,
’ὔὕ Lal ,
φθίνει θυραῖος ἠπίας χειρὸς πόθῳ.
+ 5 5 lal
οὔτοι yap ἔλαθόν σ᾽, ὦ Pir’, ὡς πῆξάν μ᾽ ἔχει
10 ἐτῶν δυσοίστων τὀυπικείμενον βάρος,
oy > -
ὥσθ᾽, οἷον ἀμπνέων τις ἐκσωθεὶς ἅλός,
ων » ans <> a ΄σ Lal
τῆς ἀλλοτ᾽ ἤλλαξ᾽ ἡδονῆς κῶμον κακῶν:
we a 3 9 & ὦ
ἔρως διῆλθε᾽ μ᾽, ὥσπερ ἐξ ὕπνου βρέφος
ἐγερτὶ δηξίθυμον αἰάζων βοήν,
- aug) 15 >
15 εἶτ᾽ εὔφρον᾽ ἐκστὰς ἐλπίδ᾽ αὐθάδης φόβος,
ἐλπίς τ᾽ ἄνελπις ἄφοβον οἰκόθεν φόβον
(ΓΑ πὸ Ὁ) 3 / ΕΣ τα ἀντ
αὑτῆς καταγνοῦσ - ησθόμην δ᾽ ὃ δύσμορος
΄ “ @ 5 » ‘I ἊΝ ’,ὔὕ
πάλαι παρηβῶν, εἶτ᾽ ἀπανδρωθεὶς μάτην,
καὶ πνεῦμα θεῖον διὰ φρενὸς μάτην πνεών,
20 μάτην τε σοφίαν εἰς ἄκραν ἀφιγμένος:
Ν , SOs Ly ΄, ε , Φ rn
καὶ πάνθ᾽, ὅσ᾽ ἀβάτους ἡυρόμην ὕλας πατῶν,
Ων , > “ > > Ν > / is
καὶ πάνθ᾽, ὅσ᾽ αὐτὸς ἐξέθρεψα, πάντα τε
cr Ate a eres, > cs ,
ταῖς σαῖς ὅσ᾽ ἐξήνθησεν ἐν ξυνουσίαις,
a Ν “-“ a 32a /
φεῦ, λυγρὰ κεῖνα viv μ᾽ ὁρῶ λωτίσματα
bo
Ol
νέκυν περιστέλλοντα καὶ λεχῶν ἔπι
cae 4 / \ ‘\ ‘4
ἐνὸς κύτους φόρημα πρὸς ταφὴν μίαν.
W. T. Lenprum.
OBITUARY.
HENRY S. FRIEZE, LL.D.
Professor of the Latin Language and Literature in the University of Michigan,
Died 7th Dec. 1889.
Tn the death of Professor Frieze, not only
the University which he served with dis-
tinction for more than thirty years, but
also the wider circle of classical scholars
aaa this country has suffered a great
OSS.
He was born in Boston, Mass., Sept. 15th,
1817. In early youth his family removed to
Rhode Island, and he was prepared for
college in a private school at Newport. In
1841 he was graduated from Brown Univer-
sity at the head of his class. At once after
K 2
132
graduation he accepted a tutorship in the
University, which position he held for three
years, when he became one of the principals
of the University Grammar School in Prov-
idence. In 1854 he was called to the chair
of Latin in the University of Michigan. In
connection with Dr. James R. Boise, who
was then Professor of Greek, he placed
classical studies upon such a broad and sure
basis in the young University that, in spite
of the βαναυσία spirit of our day and the strong
trend toward scientific studies, the ancient
classics have well maintained their position
in the college curriculum.
Professor Frieze from the first gave to his
teachings what may be called the flavour of
a noble realism. As he used to say, he was
more desirous that his pupils should be
Romans than that they should be Latinists.
It was his constant aim to imbue his students
with an interest in the old Roman life and
literature. Hence from the beginning he
accompanied his reading of texts with lec-
tures on Roman history and antiquities.
This same spirit became evident in his edi-
torial work. His first edition of the Aeneid,
which appeared in 1860, was characterized
by fine literary sense, keen insight into the
life and thought of the period portrayed by
the poet as wellas of the Augustan age, and
poetic appreciation of the author. It was
one of the earliest text-books published in
this country to give students the impression
that the masterpieces of ancient literature
were not written chiefly for the purpose of
being the vehicle of grammatical learning ;
not that Professor Frieze ignored the drier
details of linguistic study, but rather that
he placed the emphasis upon the literary
qualities and the subject-matter of the au-
thor in hand.
He had no taste for the pursuit of textual
criticism, and was but little interested in the
details of purely philological science. But he
had rare good judgment and an almost in-
tuitive perception in matters of style and
interpretation.
In 1882 he revised his first edition of
the Aeneid. The following year he edited
the complete works of Vergil, with notes
and dictionary, and a separate edition of the
Georgics, Bucolics, and the first six books of
THE CLASSICAL REVIEW.
the Aeneid. The entire work was revised
again in 1887, soas to get the benefit of the
most recent studies in Vergil. In 1865
appeared the first edition of the 10th and 12th
books of Quintilian, based upon the labours
of Spalding, Zumpt, and Bonnell. This au-
thor had never before been regularly read in
an American college, so far as we know, and
his value as a field of discipline in literary
criticism was first recognized in this country
by Prof. Frieze. The edition was greatly
improved by a revision made in 1888, which
was the last editorial work of note he
accomplished.
By years of study as well as by native
endowment, Dr. Frieze was an enthusiastic
lover of fine art and became a recognized
authority in art criticism. A remarkably
fine musician, a devoted student of architec-
ture, sculpture and painting, art studies
grew to be in his later years his most
absorbing pursuit. For the last ten years
he gave courses of lectures on the history
of art. Under his fostering care there grew
up large collections of casts, photographs,
coins and engravings, and in 1886 he put
forth a charming biography of Giovanni
Dupré together with a translation of two
dialogues on art from the Italian of Augusto
Conti.
Professor Frieze was twice called to admin-
ister the affairs of the University as acting-
president. He was the originator of many
of the plans and innovations that have made
it a leader in educational movements. Es-
pecially noteworthy are these: the estab-
lishing of the organic relation between the
University and the High Schools, by which
graduates from approved High Schools are
admitted into the University without ex-
amination ; the development of ‘the elective
system ’and of graduate courses of study ;
the introduction of the scientific study of
music into the regular curriculum.
In his personality Professor Frieze was
one of the most delightful of men, ‘a pure
and beautiful soul’ to know whom was a
benediction. To the Greek ideal of the 6 καλὸς
κἀγαθός he added the Christian graces of
charity and humility.
M. L. D’Ooce.
Michigan University.
ARCH AOLOGY.
ATHENA PARTHENOS.
Ir would be interesting to know on what
principle the Athenians in the early part of
the fourth century B.c. employed as a device,
representative of the city, at one time the
archaic figure of Athena Polias, at another
the late type of Athena Parthenos, created
THE CLASSICAL REVIEW.
by Pheidias. On the Panathenaic prize-
vases of that date the Athena is of the
archaic form, while on a series of contempo-
rary reliefs in marble which served as the
headings of treaties or alliances we have
the Parthenos. In these reliefs Athens in
the person of Athena takes the hand of
some other state. Athena is not so much
the goddess as a symbol of the power of
Athens, easily recognizable from the identity
of her form with that of the famous statue
by Pheidias.
One would suppose that the Parthenos
had been chosen for the device of the city
on occasions when it was merely a question
of some act of peace, as in these alliances
or in the honorary decrees where she ap-
pears in much the same type. But an ex-
ception to this view of the matter is fur-
nished by an engraved gem lately acquired
by the British Museum. The gem is of
sard and in the form of a scaraboid mounted
ona silver ring. It was found in Cyprus,
and from the style of the engraving it
belongs to a date shortly after 400 B.c. The
subject consists of a figure of Athena stand-
ing to the front, wearing her helmet and
aegis. At her left side are the shield and
spear, the shield resting on the ground. At
her right side is the serpent associated with
her worship on the Acropolis of Athens. So
far she answers perfectly to the Parthenos
of Pheidias, and indeed the engraver, to get
the resemblance to the face of the Parthenos
more distinct, has enlarged the head out of
proportion to the rest of the figure. But
whereas the Parthenos held out a figure of
Victory in her right hand, the Athena on
the gem holds out in her right hand the
acrostolion or ornament on the stern of a
ship, and this, as we know from ancient
records, was the recognised emblem of a
naval victory. It was not the only emblem
of a naval victory, because Herodotus (viii.
122) tells us that the Aeginetans recorded
their share in the battle of Salamis by
setting up at Delphi a bronze mast with
three golden stars at the top, ἐπὶ τῆς γωνίης,
whatever that may precisely mean. But the
usual emblem was an acrostolion or acro-
terion as it was also called. The Athena on
the gem must therefore refer to some naval
victory.
On a silver coin of Cyprus, bearing a
Cypriote inscription which is supposed to
refer to a king of Citium, Demonicos (B.c.
400-368), we see a figure of Athena very
much of the same type as on our gem. But
on the coin she sits on the prow of a ship,
holding out the acrostolion in her right
133
hand. This coin is engraved as a vignette
to Kekulé’s Keliefs an der Balustrade der
Athena Nike, and whether rightly or not as-
signed to Citiuin, it clearly commemorates a
naval battle in which Athens had directly or
indirectly aided one of the Cypriote towns, as
in fact she was constantly doing in the be-
ginning of the fourth century B.c., by sending
such ships as she could spare. [ is possible
that our new gem belongs to this same cate-
gory of memorials of a naval victory in
which Athens had aided some one of the
Cypriote towns, in which case the Athena
would play the same ré/e as on the reliefs of
the alliances and honorary decrees. It may
be objected, however, that this view of the
matter gives too much importance to the
fact of the gem having been found in Cyprus.
So small an object could well have been im-
ported from Athens, where it may have been
engraved with reference to some purely
Athenian victory at sea. These are possi-
bilities. But the character of the art is
against them. The gem, though extremely
beautiful and attractive, shows on a close in-
spection that it owes much of its attract-
iveness to the original from which it has
been copied. In itself it fails in several
points where an Athenian engraver of the
time would hardly have failed.
In his interesting paper on ‘ Personifica-
tions of Cities in Greek Art,’ which appeared
in the Hellenic Journal for 1888, Professor
Gardner dealt briefly (p. 59) with certain
instances in which a naval victory was com-
memorated by a figure holding out an acro-
stolion. There was the figure personifying
Salamis which Panaenos painted on the
barrier of the throne of Zeus at Olympia,
and there was the colossal statue which the
Greeks set up at Delphi in memory of that
same battle. Herodotus (viii. 121) says that
the Greeks sent to Delphia quantity of the
spoils, adding, ἐκ τῶν ἐγένετο ἀνδριὰς ἔχων ἐν
τῇ χερὶ ἀκρωτήριον νεός. Professor Gardner
concludes that this statue was a personitica-
tion of Salamis, just such as Panaenos after-
wards painted at Olympia. But that ex-
planation, though perfectly reasonable in
itself, is not without a difficulty. Pausanias
(x. 14, 3) says that the Greeks set up at
Delphi a statue of Apollo to commemorate
their deeds at Cape Artemision and at
Salamis, ἀπὸ ἔργων τῶν ἐν ταῖς ναυσὶν ἐπί τε
᾿Αρτεμισίῳ καὶ ἐν Σαλαμῖνι. It seems to me
that this must be the statue of which He-
rodotus spoke, and if that is so, then between
these two writers we ought to think of the
statue as an Apollo holding out an acrostolion
in one hand. That does not seem to me an
-
134
improbable conception. It would be a
counterpart to the Athena on our gem.
A. S. Murray.
ARION.
Is the poet Arion of Methymna a real
person? He has doubtless been frequently
dissolved, but he has persistently recrystal-
lized, for I find him cited in the most recent
books as the inventor of the dithyramb.
The only evidence for his existence which
need be considered is that of Herodotus,
who writes as follows: ‘Periander was
tyrant of Corinth. The most wonderful
thing that the Corinthians tell about him is
the story of how Arion of Methymna, the
first κιθαρῳδός of his time, and as far as is
known the inventor and namer of the dithy-
ramb which he produced in Corinth, was
carried to Taenarus on a dolphin’s back.
(The Lesbians confirm this).’ Such seems to
me to be the sense of the passage. It is
improbable that Herodotus wishes us to
understand that the Lesbians are also his
authorities for the statement that Arion
produced the first dithyramb in Corinth. He
then goes on to tell the well-known tale,
how Arion undertook a tour in Italy and
Sicily, the America of his time, with the
object of making money, how on his return
he took passage in a Corinthian ship at
Tarentum, and the rest. At the end he
again states that the story is so told both in
Lesbos and at Corinth, and then he adds:
‘There is at Taenarus a little bronze
offering made by Arion, a man riding on a
dolphin.’ That the identical story with all
its poetry and all its Corinthian touches
was told as a Volksmérchen both at Corinth
and in Lesbos is impossible.. All we can
get from Herodotus is: ‘This is what the
Corinthians narrate and the same story is
told in Lesbos.’ He gives us the whole in
equal good faith, but it has been tacitly or
overtly resolved into a historical and mythi-
cal portion: historical, Arion of Methymna
the dithyrambic poet at Periander’s court ;
mythical, his shipwreck and ride on the
dolphin. I believe that this division is un-
justifiable and that Arion is just as mythi-
cal, or in fact the same, as his namesake the
gifted horse.
That Arion was a local divinity, at or near
Taenarus, not unconnected with horses, is
established by the evidence of an old Spartan
inscription (1.4.4. No, 79), from which we
learn that there was in Laconia a goddess
THE CLASSICAL REVIEW.
called ’Apiovria in whose ‘temenos’ hippic
contests were held. ᾿Αριοντία is unquestion-
ably derived from ’Apiwy. Rohl (ad loc.)
conjectures that she may be the same as the
᾿Δφροδίτη ’Apeia of Sparta, but I have little
hesitation in pronouncing her to be a Demeter
corresponding very closely to the Demeter
Erinys of Thelpusa, the mother of the horse
Arion, and to the analogous Demeter
Melaina of Phigaleia, whose horse-headed idol
held a dolphin in its hand. The statuette
at 'Taenarus, apparently inscribed with the
name of Arion, represented, not a horse, but
a man riding on a dolphin, and very
probably (although Herodotus does not say
so) holding a lyre lke the Phalanthus on
the coins of Brentesium.t We should have
called it a statuette of Apollo Delphinios.
The very close connection of Apollo Del-
phinios with Poseidon Hippios and _ his
horse-children can be so widely illustrated,
that its complete demonstration is impossible
here. The conclusions to which I have been
led are these: —
The oldest oracles in Greece were water-
oracles aud belonged to Poseidon.
The horse and dolphin were manifesta-
tions of Poseidon especially in his oracular
character.
Apollo when he succeeded to Poseidon as _
an oracle-god inherited these emblems.
It may be well, as we are here concerned
with Taenarus, to give some facts concern-
ing it which illustrate these propositions,
although the legends of Delphi and other
places are more conclusive. There is at
Taenarus a trace of a disused Poseidonian
oracle, a magic well in which people saw, not
indeed the future, but the ships and the
harbour (Paus. IIL 25). Associated with
Taenarus (see Miiller, Orchomenos, p. 309) is
Euphemus the son of Poseidon and Europa
(another form of Demeter Erinys), whose
mantic character is indicated by his gene-
alogy, by his name, and by the story in
Pindar’s fourth Pythian that he received
from the Triton Eurypylos? the prophetic
clod. Apollo Delphinios in the Homeric
hymn comes to Taenarus in his progress to
Delphi, a progress which consists of a
pilgrimage to, and annexation of, various
Poseidonian sanctuaries, and we find
1 Phalanthus also appears in hand-books as a
historical person, but he has been recently dis-
posed of by Studniezka (Kyrene, Appendix 1.)
May there be something horsey in his name and
may he be connected with the κατὰ πρύμναν ἥρως of
Phalteron ?
2 Eurypylus himself is a son of Poseidon and
Celaeno—an oracular Poseidon who like so many
others has found his ultimate home in the sea,
THE CLASSICAL REVIEW.
Apollo’s rights here sanctioned by the old
distich :—
σόν τοι Δῆλόν τε Καλαύρειάν τε νέμεσθαι
ΠΠυθώ τ᾽ ἠγαθέην καὶ Tatvapoy ἠνεμοέσσαν
(Paus. IT. 33).
So far the association of Arion of
Methymna with the Arion of Taenarus is to
some extent accounted for. What makes
me doubt the historical character of the
former is that at Methymna we find the
Apollo-Delphinios legend, and we find it
there in two forms which, taken together,
admirably illustrate the connection of Apollo
Delphinios with Poseidon Hippios. The
local name of the god or hero is here
Enalus. According to the story told by
Plutarch (Seip. Conv. vii. 20) he was in
love with the daughter of Smintheus (Apollo
Smintheus, probably an oracle god). When
she was cast into the sea he followed her,
but was saved by a dolphin and brought to
land, where he was worshipped. Anticleides
(cf. Athen. XI. 466) does not speak of the
dolphin, but relates how Enalus was drowned
and how many years afterwards he appeared
from the sea and told that he shepherded
the horses of Poseidon. Here then, if we
suppose Herodotus’ testimony to imply that
the dolphin story was told in Lesbos abouta
person called Arion of Methymna, we have
good reasons for not dissociating this Arion
from the horse Arion.
It may be the result of chance that we
know of no other mythical Arion at
Methymna. At Miletus, with its neigh-
bouring spring oracle of Branchidae and with
its dolphin-rider Coeranus, the name ἐδ
found as that of an old king connected by
legend with Hesione, whose story is similar
to that of the daughter of Smintheus.
The dithyramb (whatever it was) may or
may not have been invented at Corinth. I
think from the preceding considerations that
it is dangerous to state as a historical fact
that it was first produced there by a native
of Methymna called Arion.
W. RB. Paton.
ACQUISITIONS OF BRITISH MUSEUM.
1. A chisel, axe and knife in bronze, from a small
island Suria (probably the ancient Nisyros), north of
Carpathos,
2. Three vases from primitive (‘island’) graves in
Antiparos. ‘Two are of terracotta, roughly made with-
out the wheel, and decorated with vertical bands of
herringbone pattern incised: the one is in the form
of a pyxis 4 in. high, with holes pierced in the lid for
fastening to two corresponding loops pinched outin the
sides: the other is like a small lebes upon a foot,
narrowing at the neck. The third vase is of marble
(lychnites), and is a goblet of simple form, almost
cylindrical, but narrowing towards the foot: on
135
opposite sides are two slight projections, pierced as
if for attachment, possibly of a lid now missing.
3. A series of objects excavated in 1889 by Mr. and
Mrs. Theodore Bent in the island of Bahrein in the
Persian Gulf—a description of this island was given
by Mr. Bent in the Classical Review ante vol. iii. p. 420.
They consist of :—
(i.) Fragments of ivory, including one piece carved
with the hoof of a bull and others incised with rude
patterns.
(ii.) Fragments of ostrich egos which seem to have
been decorated with colour and incised designs.
(iii.) Fragments of bronze and pottery.
A fuller account of the expedition and its results is
printed in the Proceedings of the Royal Geographical
Society, January 1890.
4, A series of marbles which were brought to this
country in the beginning of the present century by
Lord Spencer (see Michaelis, dnc. Marbles, p. 716)
who then resided at Wimbledon Park. They consist
of :-—
(i.) Marble vase cylindrical in form, sculptured
with eight female figures in relief, and inscribed with
a dedication as follows: Ζώπυρος Ζωπύρου τὸν οἶκον
‘Eorla καὶ τῷ Δάμῳ in letters of a late period : see
Michaelis bid.
(ii.) Marble drum of a fluted column—Michaelis
thought that the drum and sculpture combined might
have belonged to a columna caelata—this is impossible
on the score of size.
(iii.) Square block of granite.
5. Four vases of lychnites and four of terracotta
hand-made and roughly baked: from primitive
(island) graves in Antiparos.
6. Three objects from the collection of the Earl of
Carlisle ;
(i.) The krater signed by Python, described in
Klein Meisters.? p. 210, see also Michaelis in Hellenic
Journal 1885, p. 40. These publications render a
detailed description unnecessary here: it is sufficient
to state that the obverse side represents the apotheosis
of Alkmena: she sits on an altar (7) at the back of a
burning pyre, in an attitude of appeal to Z:-us who is
half seen in an upper plane on the left. The pyre is
being lighted with torches on either side by Amphi-
tryon and Antenor. Above Alkmene is the arch of
the sky represented like a rainbow, within which the
entire space is painted with white dots representing
mist (ἢ. From the upper part of this arch two
Hyades half seen pour water upon the pyre from their
hydriae—and Zeus has launched two thunderbolts
which lie on either side of the pyre. The figure of
Zeus is balanced by that of Eos (AQ) half seen.
Each of the figures in the design has the name incised
above the head.
On the reverse is Dionysos (youthful) ; moving
between two dancing maenads. Above him is half
seen a figure like himself: and on either side in this
upper plane a Satyr and a curious type of the goat-
legeed Pan, half seen, with uplifted arms.
The obverse does not seem to have been retouched
as much as Michaelis suggests: the altar(?) on the
obverse has been restored as a sarcophagus with pillars,
and the pyre and thunderbolts seem retouched, but
this is all. The name of Amphitryon terminates in
an H instead of an N : and Python’s name has © in-
stead of © as Klein gives it.
(ii.) and (iii.) Two large phalerac of chaleedony in
excellent condition. The one (4 in. high.) represents
the upper half of a Roman boy. The other (3 in.
high) is smaller, but the boy has a topknot and
carries a bunch of grapes resting against his chest.
Crcin SMITH.
136
Thr ErHNoLocic AFFINITIES OF THE ANCIENT
Erruscans.—In Mr. Brinton’s paper read before the
American Philosophical Society we are invited to a
fresh argument on the much disputed subject of the
origin of the ancient Etruscans. We are asked by
him to consider the possibility of their early home
having been in the north of Africa, in a portion of
the province of Algiers, known as ‘La Grande
Kabylie,’ and inhabited by the Kabyles, ‘the most
direct descendants of the ancient Libyans.’ 1
His general conclusions may be briefly summa-
rized as follows :—
1. Geographical Position. In common with the
general belief, he holds that they were intruders on
Italian soil, and that the unanimous testimony of
antiquity points to their having come from the
south. By tradition and religious custom, the city
of Tarquinii, the modern Corneto, a few miles north
of Civita Vecchia, has been generally fixed upon as
their earliest permanent settiement. Here it was
that the priests and soothsayers resorted to perfect
themselves in the ‘Etruscan discipline,’ and here
also their hero god Tages, who taught them the art
of divination and the nobler possibilities of life, came
into being.
2. Physical Traits. The notion of their having
been short in stature, thick-set and dark is, in his
opinion, partly taken from representations on cinerary
urns and partly from the writings of two late Roman
poets—Virgil and Catullus. He, however, draws at-
tention to the fact that actual measurements, made
by Italian anatomists, of many skeletons show them
to have been an unusually tall race, the average
being nearly five feet nine inches. This is but slightly
above the average height of the Kabyles. He further
points out that cranial measurements show almost
identically the same result, and that the skulls in
both instances are of the long type—dolicocephalic.
3. Culture Elements. That the ancient Etruscans
came from a distance he considers evidenced by
their not having apparently possessed, originally,
the higher arts of life, and by the fact that their
alphabet appears to have been derived directly from
the Greek : also by their having possessed character-
istics (such, for example, as the position of social
equality assigned to women which still prevails
amongst the Kabyles, notwithstanding their con-
nection with Mohammedanism) not pertaining to the
nations in the midst of which they settled. He also
shows that the main feature of their political institu-
tions, confederation, differed from that of their neigh-
bours, and was in common with that prevailing in
North Libya, as evidenced by the very meaning of
the word Kabyle—the Arabic q’bail—confederation.
4, Language. The author, in instituting a com-
parison between the Etruscan and Lybian or Kabyle
languages, points out that difficulties are occasioned
by an insufficient knowledge of the Lybian tongues,
ancient and modern, and by the fact that the prin-
cipal remains of the Etruscan language consist of
short sepulchral inscriptions. He agrees with two
of the best authorities in assigning a masculine and
feminine form to the nouns—in this resembling the
Libyan. Also that, although conjugations and de-
clensions have not been finally defined, a terminal s
added to words is believed to be a sign of the genitive
or possessive case, and in the Libyan dialect the
terminal s is found with the same possessive sig-
nification. Many words in the two languages are
1 The suggestion made by Dr. Brinton of the
Libyan origin of the Etruscans is not new, although
he seems to regard it as such. Vide Dennis’ Cities
and Cemeteries of Etruria, introduction vol. i., page
xxxix., third edition.
THE CLASSICAL REVIEW.
compared, but want of space prevents more than one
or two being given.
‘ Aesar, a god, may be derived from the Libyan
asr, light; esan, lightning. Lightning is the con-
stant accompaniment of the chief Etruscan deity.’
Ath, man; ara, descendants ; atar, family. These
words frequently recur on sepulchral inscriptions and
constitute one of the strongest points of evidence, for
in ail Libyan tribes the syllable at, ar, ath, or wit 15
the sign of tribal kinship.
᾿Αφροδίτη ὡπλισμένη. ---΄π the first part of the
‘Numismatic Commentary on Pausanias’ of MM.
Imhoof-Blumer and Perey Gardner (J.H.S. vol. VI.
p. 75) we find as many as seven autotypes of coins (i.)
of Corinth bearing types of an Aphrodite which are
manifestly derived from the same original and, as more
than one coin indicates, froma temple statue and from
a temple in Acro-Corinthus. One is marked autono-
mous, but all the others belong to Hadrian and later
emperors. An Eros is in attendance on most, in
other cases a second appears (cf. Pindar’s Corinthian
Scholion, ματέρ᾽ ἐρώτων οὐρανίαν). We cannot doubt
that these coins reproduce the type better or worse of
the statue of Aphrodite which Pausanias saw on
ascending Acro-corinthus. ᾿Ανελθοῦσι δὲ ἐς τὸν
᾿Ακροκόρινθον ναός ἐστιν ᾿Αφρυδίτης, ἀγάλματα δὲ αὐτή
τε ὡπλισμένη καὶ Ἥλιος καὶ Ἔρως ἔχων τόξον, Xi,
i, al.
The epithet here given to Aphrodite misled Lenor-
mant into regarding the helmeted head on the early
autonomous coins of Corinth as that of the armed
Aphrodite ; this statue was evidently not helmeted,
and there is no record to appeal to beyond the passage
of Pausanias. But it is as well to observe further
that the coins are proof of the sense in which
Pausanias used ὡπλισμένη, being simply ‘equipped
with a shield,’ not armed generally. See Liddell and
Scott for the later limitation of ὅπλον to the shield
alone, exclusively not only of offensive weapons, but .
even of cuirass, helmet, ὅθ. Aphrodite stands un-
draped to the hips and holding up a shield—that of
Ares of course—as a mirror. It is rather adventurous
to lay down that ‘this is a motive natural to Roman
rather than to Greek art, and we may be almost sure
that the statue does not date from an earlier period
than Julius Caesar: indeed to his time it would be
peculiarly appropriate, considering his descent and
pretensions.’ We are scarcely entitled to limit the
range of motives in Greek art of the best time. The
Scholion of Pindar warns us that the worship of the
goddess at Corinth was at its height before Rome
was heard of. The figure on the coin represents, in
pose of limbs and turn of body, the identical type of
the Venus of Melos and the bronze Victory of
Brescia : only that the artist—so to call him—evaded
the difficulty of showing the shield resting on the
mid-thigh by calling on the goddess—more absurdly
than audaciously—to sustain it freely without a
rest.
As the Brescian Victory is not contemplating her
reflection but inscribing names of warriors on the
shield, it is quite appropriately that she has the
additional drapery of a tunic.
W. Warkiss Lioyp.
Rémische Mittheilungen. 1889, part 1. Rome.
1. Mau; excavations in Pompeii 1886—88 ; a de-
tailed description of the houses in insula ix. 7, con-
sisting principally of shops, taverns etc. : plan. 2.
Wolters : notes on Greek portrait sculpture : (i) Se-
leukos Nikator and (ii) Ptolemy Soter: (i) a bronze
bust from the villa at Herculaneum, Comparetti La
THE CLASSICAL REVIEW.
Villa Ere. Taf. 10, I p. 264. 19: (ii.) marble bust in
the Louvre no 457: two plates, two cuts. 3. Hiil-
sen: antiquities of the Monte Citorio ; (i) the Colon-
na del Divo Pio: (ii) ancient building excavated in
1703 under the casa della missione (from a MS. with
several cuts in text). 4. Petersen: the Hera of Al-
kamenes : Overbeck Kwnstm. iii. p. 461 has collected
the copies in statuary of this type: reproductions of
it are found in reliefs of about B.c. 400: three cuts.
Reports of meetings with summaries of some papers
read : esp. (see p. 85) Petersen on an archaic statue
of an Amazon shooting and on Athena among the
nine Muses on the frieze of the forum of Nerva (with
three cuts).
Cs:
The same. 1889, part 2. Rome.
1. Gamurrini; the matrimonium IJtalicum ; pub-
lishing an early Etruscan relief from a tomb at Chiusi
representing this rite: plate. 2. Mau; excava-
tions in Pompeii 1886—88 (continued). 3. Winne-
feld : antiquities of Alatri (anc. Aletrium, a city of
the Hernici with pre-Roman fortifications), excavated
in this year: (i) the architectural remains, (ii) the
temple, (iii) a Roman inscription of the time of the
Gracchi, recording the public works carried out by L.
Betilienus : seventeen cuts, two plates. 4. Schneider:
on the Attic vase-painters; publishing the Glau-
kytes cup in B.M. (B 364.), the Anakles (?) cup
(Klein AMeisters.” p. 76, no. 3) and the cup (Klein,
ibid. p. 75 note) which he reads as by Tleson : plate,
two cuts. 5. Wernicke: bronzes of Epidauros, be-
longing to Count Tyskiewicz: specially interesting
is a statuette inscribed Ὑβρίστας ἐποίηξε ; four cuts.
6. Mommsen : miscellanea epigraphica. 7. Hiilsen:
the cestus of ancient boxers: four cuts. 8. Dessau:
epigraphical notes. Reports of meetings: two cuts.
Cas:
Arch.-Epig. Mittheilungen aus Oesterreich. 1888,
part. 2. Vienna.
1. Klein: studies in the history of Greek Paint-
ing (continued:) II. the Helladic and Asiatic
schools: a long article, full of important suggestions
and emendations. 2. Weinberger: on the honorary
decreee from Tomi (see ante, vol. xi. p. 41, no. 55):
emendations and suggested restoration of the whole.
3. v. Premerstein: on the inscription CZ, Z. iii. 4037.
4. The same: note on ante, xi. p. 240. 5. v. Do-
maszewski: bronze plate with bronze figures at-
tached, found in the Siebenbiirgen ; forming the de-
coration of a Roman horse trapping: plate. 6. Hau-
ser: report of excavations in Carnuntum: with 7.
Schmidel : the find in the amphitheatre there ; and
8. Bormann the inscriptions : tour plates, and three
cuts. 9. Hiilsen ; a Roman street in Servia. 10.
Kubitschek : the so-called ‘Roman sarcophagus’ in
Gumpoldskirchen : this trough, the only evidence of
Roman conquest in this neighbourhood, has two in-
scriptions which are not ancient: the trough is not
ancient either. 11. Bormann: the ancient inserip-
tions at Wodena (Edessa).
137
Zeitschrift fiir Numismatik (Berlin), Bd. xvii. Heft
1. 1890.
A. Loebbecke. ‘Griechische Miinzen aus meiner
Sammlung. iv.’ (Plates I. II). Among these may
be noticed :—Arnae in Macedonia. The coins attri-
buted to this place by Millingen and Head would
seem rather to be of some Italian city. Loebbecke
obtained six specimens from Naples together with a
number of coins of Italy. Dicaea (Thrace). Archaic
tetradrachm with head of Herakles ; a new denomi-
nation. Sicyon. Rare gold coins of the fourth
century B.C. with typesof Apollo and Dove. Loebbecke
suggests some good reasons for believing in their
genuineness. Phaestus. An interesting variety of
the staters representing Talos the winged guardian of
Crete, accompanied by his dog. Dardanus. Elec-
trum stater, type, ‘cock with mussel in mouth.’ The
‘mussel’ however is more like a fish’s head such as
appears on electrum coins of Cyzicus. Ceretape
(Phrygia). Coin of Caracalla with his name Bassianus.
Siocharax (Phrygia). A coin of Geta with the name
of the town Siocharax, a place apparently identical
with Hierocharax.—Otto Seeck. ‘Die Miinzpolitik
Diocletians und seiner Nachfolger.’
WARWICK WROTH.
Numismatische Zeitschrift (Vienna). vol. xxi.
January—June 1889.
W. Drexler. ‘Der Isis- und Sarapis-cultus in
Kleinasien.’ A collection of inscriptions, and other
moruments, especially coins, relating to the worship of
Isis and Sarapis in Asia Minor (exclusive of the ad-
jacent islands). The paper contains much useful
material, but a summary of results and the curtail-
ment of some of the numismatic descriptions and re-
ferences would have rendered it more convenient for
consultation.—A. Markl. ‘Gewicht und Silber-
gehalt der Antoniniane von Claudius IJ. Gothicus.’
Wee
Numismatic Chronicie, Part iv. 1889.
E. Thurston. ‘On a recent discovery of Roman
coins in Southern India.’ A find of fifteen Roman
Aurei at Vinukonda (Madras Presidency). They
are coins, in good preservation, of Tiberius, Vespasian,
Domitian, Hadrian, Ant. Pius, Faustina I., M.
Aurelius, Commodus and Caracalla. —G. M. Arnold.
‘The Roman station of Vagniacae.’ Southfleet, not
Maidstone, is Vagniacae. Remarks on coins found at
Southfleet.—F. Latchmore. ‘On a find of Roman
coins near Cambridge.’ A hoard of about 2,500
coins discovered early in 1889 ‘near Cambridge’ in
two Roman jars; the coins range in date from
Gordian III. (A.p. 238) to Aurelian (A.D. 270).
There were 858 of Tetricus I. and IJ., 634 of Victor-
inus, and 411 of Postumus.—Miscellanea. C. Oman.
Coin with the name and bust of Maximianus Herculius
struck at Colchester.
Ἄν:
SUMMARIES OF PERIODICALS.
Journal of Philology. Vol. xviii. No. 36.
THis number contains several important articles on
textual criticism. W. Leaf points out the grave de-
fects in La Roche’s apparatus criticus on the JJiad,
and suggests certain criteria for determining the com-
parative value of the MSS. By the use of these he
establishes the ‘overwhelming importance’ of the
Leipzig group of MSS., as preserving an extremely
ancient tradition unknown to any of our other MSS.
E. G. Hardy gives an account of a Bodleian MS. of
198
Jerome’s Eusebian Chronicle, which strangely enough
seems to have been entirely neglected, till attention
was called to it by a German visitor about two years
ago. Yet it is at least a hundred years older than any
other MS., having been written by an intelligent and
accurate scribe within a century of Jerome’s death.
A new edition based on this MS. is being prepared by
Dr. A. Schoene. J. Armitage Robinson argues against
Dr. Koetschau, that V (the Vatican Gr. 386) is the
original of all our MSS. of Origen against Celsus,
Robinson Ellis gives an account of the Codex Moreti, a
12th century MS. of Ovid used by Heinsius. Of the
other papers the most interesting are those by H.
Nettleship on literary criticism in Latin Antiquity,
and by C. Taylor on the relation between the Shepherd
of Hermas and the Didaché. A. Platt propounds a
theory as to the origin of the lambic Trimeter ; J. P.
Postgate writes on ne prohibitive with the second
person of the present subjunctive in classical Latin ;
R. Ellis has two emendations on Lucretius and notes
on some epigrams of the Greek anthology; and H.
Nettleship closes with a note on Georg. 1. 263.
Transactions of the Cambridge Philological
Society, Vol. III. P. ii.
Notes on the Spiritus Asper. HH. D. Darbishire.—
The object of this paper is to show that the pheno-
mena of the rough breathing in Greek are much more
recular than is usually supposed. After premising
that the rough breathing is the independent develop-
ment of eaeh 1. KE. language in which it is found and
varies in its phonetic value, the writer states and illus-
trates the ordinary rules that in Greek it represents
original s, v, su, 1, stinitial. In the text-books a num-
ber of exceptions are recognised. These are to be ex-
plained as the result of analogy e.g. ἅδην on the model
of ἁνδάνω, or dissimilation e.g. ἔδαφος for ἑδαφος (σεδα-
gos) according to Grassmann’s law. But where the
breathing cannot be so explained and a rule is broken
we must regard the received etymology as mistaken.
For instance éreds must not be referred to satya
because of the smooth breathing, but must be re-
garded asa verbal of ἦμί and compared with Lat.
verus ; ἄσμενος again is not to be connected with
ἥδομαι, but referred to aroot van, cf. Lat. Venus ; ἄλτο
must be severed from G&AAoua and placed beside
ἀλείς. There are many cases where the presence or
absence of aspiration is not persistent. ‘These must
be explained either by misplacement of the aspirate,
a doubtful principle, or by assimilation to connected
words ¢.g. ὅρπηξ on the analogy of ἅρπη, or thirdly
the confusion is due to the intermixture of different
roots. Thus from the root sel we have aspirated words
ἑλίσσω, &e. ; from the root 72d unaspirated εἴλω, &e.
This leads to confusion, ¢.g. ἐλύω, ἑλύω, &e. Again
the variation of ἄνυω, aviw is to be referred to the
roots san and van ; similarly εἴργω and efpyw imply
two roots, and ἱερός (iapés), Sk. 7sivas, has been as-
similated to ἱρός from the root v7. We now come to
the question of initial F. Why is this sometimes re-
presented by the rough, sometimes by the smooth
breathing? This is to be explained by the help of
Armenian. It is generally recognised that the
initial ὁ can be distinguished from initial spirant y
c.g. 1ag- appears in Gk. ἧπαρ but yug- in Gk. ζυγόν.
Hitherto no distinction has been made between wu and
the spirant ἡ. Mr. Darbishire shows that Armenian
has sometimes g and sometimes v, w, where Latin and
Sanskrit have v. Now in Greek where initial F
answers to Armenian g we have the smooth breathing,
where it answers to Armenian v, w, we find the rough
breathing. Upon this ground then we are to assume
an original distinction between u and v and can ex-
plain the variation of the breathing in Greek. Thus
THE CLASSICAL REVIEW.
from the root wes comes ἔαστυ ἄστυ, but from the
root ves εννυμι ἕννυμι. It will be seen that the
chief points of interest in the paper are the explana-
tion of variatious of breathing by the assumption of
different roots, and the distinction of original w and »,
which is the most important and ingenious. Many
will hardly be robust enough to separate some of the
words which we have al] been taught to regard as
cognates. Thus ἕως and αὔως (4ws), and ἦμαρ and
ἡμέρα ave to be divorced. But even supposing that
some of the writer’s etymologies are not altogether
convincing the paper is a valuable and suggestive at-
tempt to bring under phonetic laws what has hitherto
been left unexplained. The representation of original
w- by Greek ὑ- he has himself to leave unaccounted
‘for. J. HE. K.
Meeting of the Philological
Society.
At the Annual General Meeting of the Society, on
Jan. 30, 1890, the President, Dr. Peile, read a paper in
which he called attention to a modification in the
latest editions of Paul’s Principien, pp. 58—60, and
Brugmann’s Greek Grammar, p. 11 (in Miiller’s
Handbuch, &c.), of the doctrine of the inyariability
of phonetic sequence. Paul distinguishes from the
examples of regular substitution of one sound for
another, others not regular which he describes as in-
terchange in certain definite cases. These are (1)
metathesis, 6.0. wasp=A.S. weeps, ψεξε σφε, where
the sounds are consecutive, or O.H.G. ezzih (now
essig)=acetum, where they are not consecutive, (2)
assimilation of two sounds not consecutive, as quin-
que for Idg. penqe, (3) dissimilation, as pelegrinus
for peregrinus ; or δρύφακτος for *Spuppaxtos, where
r is lost, or semestris for semimestris, where a syllable
nearly the same as the following syllable falls out.
Jn like manner Brugmann distinguishes gradual and
progressive substitutions of sound from metatheses
which take place by sudden transition. The writer
of the paper gave reasons for inferring that Brugmann,
though he only specifies metatheses, yet may be sup-.
posed to include the other changes given by Paul, as
cases where strict uniformity was not to be expected ;
and he pointed out that Brugmann gave a reason for
this invariability (which Paul had not), viz., that the
changes were sudden. It was then suggested that if
regularity was not to be expected in assimilation of
nonconsecutive sounds, on the ground that such as-
similation was sudden, there might be no reason to
expect it when the sounds were consecutive, because
it might be plausibly maintained that such assimila-
tions were also sudden. It is notorious that in Latin
numerous variations from the ordinary law are found
in these assimilations: 6.5. we find quondam for
*quomdam with change of m to n, but quamde with
none ; cena for cesna, but uerna for *uesna; collis
for *colnis but uolnus, where 77 remains ; porrum for
*porsuin but dorsum ; and many like ones: even if
we allow that some such variations have been plausi-
bly explained, yet many remain for which no explana-
tion has ever been offered. The writer suggested that
thoroughgoing adherents of the dogma of unvarying
phonetic sequence might find comfort in this ex-
planation of variations which in fact are found, viz.
that when change was sudden, regularity was not to
be expected. But he pointed out there was a pre-
liminary point to be settled. The most distinguished
phoneticians are not agreed whether there is such a
thing as ‘sudden’ change. Sievers (Gvundziige, p.
226, ed.%) holds that there is, and gives as an in-
stance labialism in Greek, Umbrian and Oscan.
Sweet on the other hand seems at least (1.2.8.7 §
42) to deny it. With the view of testing this point,
and so furnishing a basis for the view described
Cambridge
THE CLASSICAL REVIEW.
above, the writer gave a full list of the different kinds
of sound-change and classified them tentatively as
sudden or gradual. Upon this classification he in-
vited discussion ; and a long discussion followed.
Transactions of the American Philological
Association for 1888. Vol. XIX. Boston,
1889.
THE Lex Curiata de Imperio, W. F. Allen. Against
Mommsen’s view (Romisches Staatsrecht, i. 52) that
this law was merely a formal recognition of a power
already vested in a magistrate, the writer holds it to
have been a substantial grant of power by a body
different from that which elected (¢f. modern bicam-
eral legislatures). An emergency prior to its enact-
ment would be met in some practical way, as in the
case of a provincial army deprived of its leader, or (as
military power was vested in the consuls) through the
formula videant consules, ete., though this special
exercise of the imperiwm can scarcely have included
holding the comitia centuriata. Mommsen cites as
examples to prove his interpretation the consul
Flaminius B.c. 217, the consuls of B.c. 49, and Appius
Claudius, consul B.c. 54; but other explanations are
offered. Moreover the importance of the law is
strongly stated by Livy (v. 52, 15), by Cicero (leq.
agr. 11. 11 and 12) and by Dio Cassius (39, 19).—On
the Impersonal Verbs, Julius Goebel. A discussion
of the various theories (1) that a subject is contained
in the impersonals (Ueberweg, Lotze, Wundt, etc.),
(2) that there is no subject (Herbart, Trendelenburg,
Grimm, Benfey, etc.) and (3) that of Paul and others
who hold an intermediate position, distinguishing
between the logical and psychological subject. The
writer supplements Sigwart’s discussion (Die Imper-
sonalien ; eine logische Untersuchung) made from the
standpoint of logic, by considerations drawn from
linguistics.—On the Authorship of the Cynicus of
Lucian, J. Bridge. An endeavour to show that
Fritzsche is wrong in stating (II. 2, p. 235f.) that
the same man could not have written the Fugitivi
and the Cynicus, and to prove Lucian to be the author
of the latter. The same man could inveigh against
the Cynics of the time (to him, false Cynics) as in the
Fugitivi, and defend Cynicism as in the Cynicus. Du
Soul’s argument that the words κόμην ἔχειν are not
Lucian’s since Cynics in his day were ἐν χρῷ κεκαρ-
μένοι is unfounded ; for in his time the Cynics wore
long hair and beards, ef. Tatian, Ad. Graecos c. 25 ;
Dio Chrys. Or. 72; Lucian, Peregrinus 15. The
only exception Cantharus (fugitivi 15) had in part
adopted Stoic dress. The peculiar repetition of the
first word in a clause may have been in imitation of
Dio Chrysostom (cf. Or. 72 with the Cynicus).—The
other papers are out of our department.
In the Report of the Annual Session are summaries
of the following papers: The Changes in the Roman
Constitution proposed by Cicero (Legg. iii. 8, 6—5, 12),
W. A Merrill.— The Leaacy of the Syrian Scribes (the
President’s address), J. H. Hall, showing the import-
ance ofa knowledge of Syriac literature in the study of
the transmission of the Bible, the Greek and Latin
classics, the Greek fathers, etc., especially in matters
which the ordinary Hellenist little suspects.— The
Cure Inscriptions from Epidaurus, J. R. Wheeler.—
Goethe's Homeric Studies, G. M. Richardson.—Vola-
pik and the law of Least Effort, ¥. A. March.—
Theories of English Verse, J. C. Parsons-—A Con-
sideration of the Method employed in Lighting the
Vestal Fire, M. H. Morgan.—Peculiarities of Affix in
Latinand Greek, Ο. 85. Halsey.—On the term ‘ Contami-
nation’ used in reference to the Latin Comedy, F. D.
Allen, Contaminare means not, as is usually held
after Grauert, ‘to stick’ or ‘weld together,’ but ‘ to
159
spoil ;’ and in Terence, refers to the Greek originals
and not to Latin plays. A Greek play from which
a single scene was taken, was ‘spoiled’ for
future use.—The Tripods of Hephaestus, Τὶ D. Sey-
mour. τρίπους in Hom. Σ. 373 is not a ‘ mixing
bowl’ or ‘kettle,’ but a table (¢f. Xen. An. VII. 3,
21).—Date of the Episode of Cylon in Athenian History,
J. H. Wright, setting it nearer 640 B.c. than 612 B.c.,
the usually accepted date.—4 new word, Arbitus,
F. P. Brewer. —On the Identity of words and the mis-
application of the tern. “ Cognate’ to words which are
identical, L. S. Potwin.—The Locality of the Saltus
Teutoburgiensis, W. F. Allen —Observations on the
Fourth Eclogue of Vergil, W. 8. Scarborough.
Harvard Studies
Vol. i. 1890.
1. J. B. Greenough, The Fauces of the Roman
House. The word fauces means the passage leading
from the front door to the atriwm. (1) Fauces natur-
ally means entrance, as is shown by the careful exami-
nation of the passages where itoccurs. (2) Vitruvius’s
description can refer to nothing except the front
passage, as this is always open while the others are
always closed and so cannot form a part of the archi-
tectural feature of the interior. (3) The actual houses
show the front passage agreeing with his description,
while the other is often wanting and, when present,
never corresponds to Vitruvius’s description. The
article is illustrated with a cut.
2. M. H. Morgan, Commentatio de ignis eliciendi
modis apud antiquos. In this long article are con-
sidered the various means of kindling fire known to
the ancients. In connection therewith are cited
(with comment and explanation wherever necessary)
all the passages in Greek and Roman authors, from
Homer to Suidas, which throw light on the subject.
As an introduction are considered the means of keep-
ing fire alight in the house under the ashes, in the
public buildings, and in campaigns by the πυρφόρος.
There were four principal methods of kindling fire:
(1) by the rubbing or boring of wood; (2) by the
friction of two stones ; (3) by friction of stone and
iron; (4) from the sun’s rays. Under (1) there is a
full consideration of the πυρεῖον. It provided the
commonest means of lighting fire in the classical
period, although Nos. 2 and 3 were quicker. It was
a very ancient Indo-European method, and the instru-
ment developed into a wooden bow-drill, τρύπανον,
which bored into the ἐσχάρα. The various woods
used are mentioned and the different parts of the
drill are fully considered. The troublesome and
rare word στορεύς is treated at length. (2) The use
of stones, especially pyrites and silex, is quite a com-
mon way. In Soph. Phil. 35 the word πυρεῖα, as
elsewhere often, refers to this method, and the com-
mentators and L. and §. are wrong here ; cf. 295.
(3) Stone and iron are rarely mentioned, only in
Lucretius, Pliny Elder, and Isidore. Before consider-
ing the fourth way the writer inquires into the differ-
ent kinds of kindlings and tinder, also the use of
matches, with special discussion of the words ἔσκα,
Yoana, bona, on Which Codd., Edd., and Lexx. are
very confusing. (4) Lighting from the sun. This
entailed a consideration of what was known about
reflexion, refraction, »nd plane, convex and concave
mirrors ; also the history of the words for crystal and
glass and the use of those materials. The different
sorts of burning glasses and lenses from Aristoph. to
Isidore are considered. Then follows a consideration
of the story of Archimedes burning the Roman fleet.
Finally the method of lighting the Vestal fire is dis-
cussed. This was probably kindled every year,
March 1. The writers have left us no information
in Classical Philology.
140
about the method. Asa pure flame was wanted, it
could not be got from any other fire, but in one of
the four ways mentioned above. It was probably
the same as that employed when the fire went out by
accident, and this, Festus tells us, was by boring, and
his statement is doubtless credible. On the other
hand the method described in Plut. Vit. Nun. 9, is
in itself incredible, because the lighting of fire from
the sun’s rays was a comparatively modern invention.
The passage may also be deemed an interpolation.
A passage in Julian deserves no greater confidence.
The article is illustrated with five cuts, representing
πυρεῖα and Plutarch’s method of lighting fire from the
sun.
3. W. W. Goodwin, On the origin of the construction
of ov μή with the subjunctive and the future indicative.
The expression of denial οὐ μὴ γένηται or γενήσεται
is generally explained as involving an ellipsis of the
idea of fearing : thus οὐ δέος ἐστὶ μὴ x.7,A. The pro-
hibition οὐ μὴ καταβήσει, do not come down, is gener-
ally explained as interrogative will yow not not come
down ? and the subjunctives which occur have com-
monly been changed into futures. But not all the
prohibitions eax be interrogative, nor can all the
subjunctives be changed to futures without doing
violence to the text. Nor are all cases of the 2 pers.
of the subjunctive or future with od μὴ prohibitions.
One theory should explain all cases. If οὐ in od μὴ
γένηται is an independent negative, as it should be,
the negative force of μή must be in abeyance. Note
Plato’s favourite subjunctive as a form of cautious as-
sertion, asin ph φαῦλον ἢ, which originally meant
may it not prove bad (as I fear it may), but became
softened into J suspect it may prove bad, aud then I
think it wiil prove bad, or it will probably prove bad.
The negative οὐ μὴ φαῦλον ἢ would be it will not prove
to be bad. The independent subjunctive with μή occurs
in Homer in expressions of apprehension combined with
a desire to avert the object of fear. Between Homer
and Plato there are only eight instances of this sub-
junctive, which however show the transition from
apprehension to cautious assertion. In Plato it also
expresses honest apprehension. The dramatists use
it in both senses and οὐ μὴ λάβωσί σε might mean J
am not afraid that they will seize you, being the
negative of 7 fear they may seize you, or they shall not
seize you, the negative of I suspect they will seize you.
The denial of an apprehension could easily change to
the denial of a suspicion and then the need of an
ellipsis is avoided. Attic Greek came naturally to
use the future for the subjunctive. Then later the
second person singular probably began to be used as a
prohibition.
4, W. W. Goodwin, On some disputed points in the
construction of ἔδ ει, χρῆν etc. with the infinitive.
With certain imperfects denoting obligation, pre-
priety or possibility, ἔδει, χρῆν, etc. the infinitive is
used, and the whole expression becomes a form of
potential indicative, referring to past or present time,
and generally implying the opposite of the infinitive.
Thus ἔδει σε ἐλθεῖν, you ought to have gone (but you did
not go). These imperfects may also be used with no
potential force, thus ἔδει μένειν, he had to stay. The
first of these is ‘ equivalent to the verb of the infini-
tive in the potential indicative (with ἄν), qualified by
an adverb or other expression denoting obligation,
propriety, or possibility, which expression would
stand in the relation of an unreal condition to the
verb with ay.’ ‘It is generally laid down as an abso-
lute rule that in this idiom the opposite of the infini-
tive is always implied.’ But a large number of ex-
amples show that when these potential expressions
without ἄν stand alone they always imply the oppo-
site of the infinitive, but when one of them is made
Fence with the foil.
THE CLASSICAL REVIEW.
the apodosis of an unreal condition external to itself
it may be so modified by the new condition as no
longer to imply the opposite of the infinitive. In
regard to the distinction between ἔδει &c. with the
infinitive and ἔδει ἄν &c. with the infinitive the
following rules hold good : “1, The form without ἄν is
used when the infinitive is the principal word, on
which the chief force of the expression falls, while the
leading verb is an auxiliary which we can express by
ought, might, could, or by an adverb, 2. On the
other hand, when the chief force falls on the neces-
sity, propriety, or possibility of the act, and not on
the act itself, the leading verb has ἄν, like any other
imperfect in a similar apodosis.’ Many illustrative
examples are discussed.
5. G. M. Lane, Notes on Quintilian. 1. Shall we
say divom or divim in the genitive plural? (1) There
is evidence of long 6 in coins before the first Punic
War ; also of long ἢ in the Nuceria inscription, cer-
tainly as late as 63 B.c. (2) There is evidence that the
long 6 or long ἃ had disappeared about 100 A.D.
Therefore, originally, divom, then divdim, diviim,
detim.—2. In Quint. 1, 4, 27 correct lectwm to tectum,
because /ectwm as a participle is not nominative and
has short é.—3. In Quint. 1, 4, 16 read nutrix
instead of notriz, because of MS. authority and
because Quintilian himself indicates that it isa Greek
word. Then write Culcidis for Culcides.
6. J. B. Greenough, Some Latin Etymologies. 1.
Reciprocus is a compound of trecus andt procus.—2.
Improbus. Probus, is pro+ bus and was a mercantile
word meaning A 1 or first-class. Iimprobus meant
properly not first-class and became from its mercan-
tile use a slang word of disapproval like horrid,
mean, auful &e.—3. Rudimentum means foil-prac-
tice, the practice of the raw recruit, hence the first
essays in war. It is from a (real or supposed) verb
rudio, which is from rudis foil and would mean to
4, Desidero ought to be formed
from an adjective tdesides (or-er), which would be
compounded from de and sidus, Or it may come
through the phrase de sidere. The original meaning’
of sidus was probably position, place, desidero would
mean find or mark out of place. —5. Elementwm. 111π|8-
trations are given supporting the old view from ed em
en.—9. Praemium is prae-emium (emo to take), the
special part of the spoil taken out beforehand.—7.
Deliciae, delicatus. Analogy would lead to an adjec-
tive delicus, de+licus, from which would come deli-
ciae andf delico participle delicatus. in Varro deli-
cus means a pig weaned by its mother. Dedico
would naturally mean to pet, delicatus tender and
deliciae delight.—8. Provincia is from tprovincus
meaning ‘engaged in advancing conquest,’ and pro-
vincia would be the state or condition of a person who
is provincus.
7. C. L. Smith, On egregiwm publicwm (Tae.
Ann. 111. 70. 4). He would read egregiwm publice
locum.
8. A. A. Howard, On the use of the perfect infinitive
in Latin with the force of the present. From a full
discussion of passages the following results are
obtained. ‘In early Latin the perfect infinitive with
its proper significance was used to depend on the verb
nolo or uolo in prohibitions ; but since the verb of
wishing contained the idea of futurity, the whole
clause acquired the force of a future perfect expres-
sion. Later writers, and especially the poets, trans-
ferred this use to negative clauses, not prohibitive,
containing verbs of wishing, and secondly to clauses
containing verbs like /aboro, amo, and timeo, “ Verba
der Willensrichting.’ Since these verbs contain the
idea of futurity, the present infinitive joined with
them has the force of a future, the perfect infinitive
THE CLASSICAL REVIEW.
the force of a future-perfect. The tendency of the
Latin writers to use the future-perfect for the future,
through an overstrained desire to be exact, led
them in these clauses to use the perfect infinitive
instead of the present. The poets, and especially the
elegiac poets, took advantage of the opportunity thus
offered and transferred the use to other constructions
which did not contain a verb of wishing. The
reasons for this were two: first, the present infinitive
of a large number of verbs, which they wished to use,
on account of metrical difficulties could not be used
in their verse, or could only be used under certain
restrictions ; second, the perfect infinitive of these
verbs were peculiarly adapted to the necessities of the
last half of pentameter verse. The infinitive in this
use seemed to have the force of an aorist infinitive in
Greek, and in course of time came to be used by the
poets even where the metre admitted the use of the
present infinitive.’
9. H. N. Fowler, Plutarch’s περὶ εὐθυμίας. A de-
tailed examination of passages shows that ‘only in a
few instances has it been possible to find any indica-
tion of the origin of Plutarch’s words and doctrines.’
It would appear, however, that for his introduction
and his last chapter he adapted the corresponding
parts of Democritus’s περὶ εὐθυμίας, that for his quota-
tions and anecdotes collections of such matters were
used and that even for his ethical doctrines similar
sources were employed.
10. G. M. Richardson, Vitruviana. This contains
some anomalies of syntax and style, mainly vulgar
and colloquial expressions, noted in Vitruvius; the
limitative use of a, ab; descriptive use of cum with
a noun having the force of an adjective or adverb;
aliter having a distributive force, with afque as the
connective ; copulative nec strengthened by a follow-
ing negative ; oppido, followed by quam; the 118
instances of the favourite word quemalmodum are
given in groups ; sz (in sic) in apodosis as well as pro-
tasis, a survival of parataxis ; wf with the subjunc-
tive in wishes, exhortations, and commands, showing
the origin of its use in final clauses, which were orig-
inally paratactic ; the survival in indirect questions
of the indicative which was originally used ; an un-
common use of the infinitive as predicate after esse ;
putare in the sense of ‘ intend.’
11. H. W. Haiey, The social and domestic position
of women in Aristophanes, The object of the article
is to collect the passages in Aristophanes bearing upon
this subject and some of the inferences which may be
drawn from them. It shows that the women’s estim-
ate of themselves was depreciatory, and the men’s
was low, from which we may infer that the popular
estimate of the time was similar; that husbands
usually exercised authority over their wives, that
there was lack of confidence between husband and
wife and also of conjugal affection. Unmarried
women were closely contined to the gynaeconitis and
seldom appeared in public except on religious oc-
casions. Married women did not usually go out of
their houses without some imperative reason for doing
so, though they attended and took part in numerous
festivals. There is considerable evidence that Athen-
jan women sometimes attended the theatre, but it
seems to be against the supposition that they were
present when comedies were performed. There wasa
considerable degree of freedom in connection with the
ceremonies of marriage and burial and also in the
case of those who were poor. Women of all classes
associated freely with other women. The wife was
mistress and stewardess of the house, though she did
not generally cook, caring for the children, preparing
the wool, spinning, weaving, working in flax, em-
broidering. Most women had a knowledge of sing-
141
ing and dancing, and some instruction in τὰ γράμ-
ματα, but they were ignorant of public life.
Notes. F. D. Allen. ψαῦος in Aleman (Bergk, P.LZ.G.
III? p. 77) amended to gatos (=gaos).—C./. L. I. 199
proxuma faenisicet graver’s blunder for faenisicie,
abl. of a *faenisicies (cf. facnisicia).—Schol. Ar. Ran.
13, for φορτικευομένου read φορτακευομένου (cf. φόρταξ,
porter): likewise in Suid. s.v. Avkas, read ἐφορτα-
κεύετο. The verb φορτικεύομαι should disappear from
our Lexx.—daprvw in Heracl. Tab. I. 105 refers to
partnership.—In Ar. Ran. 180 ff. the order of verses
should be 181, 182, 183, 180 (with ὠόπ, παραβαλοῦ
assigned to Xanthias).—In Hdt. VI. 57 expunge
τρίτην δὲ τῶν ἑωυτῶν. J.B. Greenough]. In Mart.
vi. 78. 31, 32 quam is interrog. G. M. Liane]. ellum
(three examples: Plaut. Bacch. 939, Ter. And. 855,
Ad. 260) from em illum.
General Index,—Index of Citations.
Deutsche Litteraturzeitung. 1889.
No. 36. Lippelt, Quaestiones biographicae. A
careful discussion of the beginnings of Greek bio-
graphical literature and of the authenticity of
Xenophon’s Agesilaos.—Schwarz, De. M. Terentii
Varronis apud sanctos patres vestigiis capita Il, A
valuable contribution to a future collection of Varro’s
fragments, although Schwarz’s arguments are by no
means always conclusive.
No. 37. Diimmler, Akademika : Beitrige zur Ge-
schichte der Socratischen Schulen. A collection of
essays to lead to ‘emancipation from Plato as histori-
cal source.’ D.’s careful investigations lead to many
valuable results.—Krebs, Zur Rection der Casus in
der spateren historischen Gracitét.-I1. Discusses the use
of the accusative after verbs like ἀπογιγνώσκω, ete.
The author sometimes overlooks the fact that certain
constructions can be traced back to a considerably
earlier period.—Bilfinger, Der biirgerliche Tag.
Against Ideler’s view, which, as B. shows, is based
only on oman authorities, the author endeavours to
prove that the Greek day began with the dawn,
while the Roman day began officially at midnight.—
Ehrenberg, Die Inselgruppe von Milos. A good
geological and geographical description.
No. 38. Plessis, Traité de Métrique grecque et latine.
Sets forth in a short and clear manner the more common
Latin and Greek metres, and will doubtless gain many
friends as a convenient introduction.—Sexti Amarcit
Galli Piosistrati Sermonum libb. IV. e cod. Dresdensi
A 167% nune primum ed. Manitius. The edition of
the sermones of this learned man, who probably
lived at the time of Henry III. in Speier is careful
and reliable ; its defects are printed out by Traube
Zeitschr. 7. deutsches Alt. xxxili, Anz. p. 195—202.
No. 39. Aug. Marx, Griechische Mérchen von
dankbaren Thieren und Verwandtes. A good con-
tribution to comparative folk-lore.—R. Sabbadini,
Studi critict sulla Exeide. An interesting and sug-
gestive book, although its positive results are but
few.
No. 40. Joh. Schmidt, Die Pluralbildungen der
indogermanischen Neutra. Like all works of Schmidt,
a most excellent and thorough investigation, full of
interesting results.—Teulffel, Studien wnd Character-
istiken zur griechischen und rémischen Litteratwr-
geschichte. 2ed. The new edition, by the author’s
son, contains the changes which T. had made in his
own copy. ‘ Horatiana,’ two reprints (of ‘ Aeschylus
Promethie und Orestie,’ 1861, and of ‘ Kritisch-
Exegetisches,’ 1878), and an ‘ Introduction to Cic. or.
pro Quinct.’ are added.—Schmidt, Abhandlungen zur
alten Geschichte. A collection of Schmidt’s disserta-
tions and essays on ancient history, heretofore scat-
tered in magazines and therefore hard to obtain.
149
No. 41. Conradi Hirsaugensis Dialogus swper
auctores sive Didascalon, ed. Schepps. Good edition
of this literary history, valuable through the infor-
mation which it gives about classical studies in the
middle ages.—Dinarchi orationes adiectis Demadis
fragmentis, ed. Blass, ed. 2. The critical apparatus
is simplified by omitting the valueless readings of
B, L, Z, M; the readings of A and N are added.
The text deviates in more thana hundred places from
that of ed. 1.—M. Terenti Varronis rerwm rustic-
arum libri 111., ed. Keil. This text edition is a sort
of supplement to K.’s larger ed. of Varro’s d. 7. 7.,
the first volume of which is a reconstruction of the
archetype, while the present ed. gives a critically
emended text with succinct apparatus.—D. Schoeffer,
De Deli insulae rebus. A valuable treatise, chiefly
based on inscriptions published in the Bull. de Corresp.
Frellén.
No. 42. Anonymus adversus aleatores und die
Briefe an Cyprian, Lucian, Celerinus, &c. Kritisch
verbessert, erlautert und...tibersetzt v. Miodoiski. The
text is decidedly improved, three MSS. (besides
the Paris) having been collated. According to M.
(against Harnack) the author of the book lived in
the second half of the third century, and took his
knowledge of the Bible and style chiefly from Cyp-
rian.— Aeschylos Orestie mit erklérenden Anmerk. v.
Weeklein. Exceedingly valuable, showing many im-
provements on Wecklein’s critical edition of 1885.—
Kauffmann, De Hygint memoria scholiis in Ciccronis
Aratum Harleyanis servata, &e. (Breslau. Abh. IIT.
4). The first part contains a good classification of
MSS.; part second, a painstaking edition of the
text.
No. 43. Hecht, Die griechische Bedeutungslehre.
An unsatisfactory appeal to philologists to pay more
attention to Greek semasiology.—Kronemberg, Minw-
ciana sive annotationes eriticae in Minucit Felicis
Octavium. Show great familiarity with F.’s style.
K. frequently defends successfully the MSS. readings.
In his own conjectures he is circumspect and careful.
—Ohnesorge, Die rémische Provinzliste von 297. I.
Endeavours to show (as it seems, successfully) that
the list goes back to Diocletian, and has very few
later additions. O. repeats Mommsen’s, Noldeke’s,
and Czawalina’s proofs, which he strengthens in some
points. The second—more important part—will
contain the proof that 297 is the year of Diocletian’s
reform of the provinces.—Pomtow, Beitrdge zur To-
pographie von Delphi. Gives some new details, but
the author might have better waited until the exca-
vations have thrown light on many obscure points.
No. 44. N. Fornelli, La Pedagogia ὁ U Insegna-
mento ckassico. One of the best modern books on the
subject. — Hermann’s Lehrbuch der griechischen
Antiquitéten. Staatsalterthiimer. 1 Abt. The
editor’s care and scholarship deserve warm recog-
nition, but the inconveniences which lie in the
nature of the re-editing of such a book are strongly
felt. — Henry Nettleship, Contributions to Latin
Lexicography. These fragments of a Latin dictionary,
which the author from want of aid was unable to
finish, contain many interesting observations and
some new glosses.
THE CLASSICAL REVIEW.
No. 45. G. Cichorius, Rom und Mytilene. Contains
three of the most interesting inscriptions discovered
by C., which throw much light on the relations between
Rome and Mytilene in the Augustan age, with C.’s
ingenious conjectures based on them.—Hruza, Ueber
das lege agere pro tutela, WH. proves convincingly
that (lege) agere pro tutela is not the representation
of a pupillus by the tutor, but is essentially identical
with the postulatio suspecti tutoris, although for the
latter there are no cogent proofs.
No. 46. Pappenheim, Der angebliche Heraklitis-
mus des Skeptikers Ainesidemos. An unsuccessful
attempt to prove that Sextus, where he attributes
Heraclitean doctrines to Ainesidemos, refers to a con-
temporary school of philosophers, who used Ai.’s
name in order to commend their doctrines. —Juvenalis
saturae erkl. v. Weidner, 2nd. ed. Many errors of
ed. I. are corrected, but their place is taken by new
ones. —Gylling, De argumenti dispositione in satt.
IX.-XVI. Juvenalis. Contains nothing that is new.
—Dirr, Das Leben Juvenals. Attempts unsuccess-
fully a reconstruction of the original Vita.—Rudolf
Lanciani, Ancient Rome in the light of recent dis-
coveries. Deserves warm commendation.
No. 47. Arnaldo Foresti, saggi sulle fonti della
epopea Greca. Is rather a physical mythology dis-
solving everything into fog, sun, and some other
natural phenomena. Full of the most fanciful etymolo-
gies.—P. Nigidii Figuli operum reliquiae coll. em.
en. Swoboda. A good collection of N.’s fragments
with critical apparatus. In the appended ‘quaes-
tiones Nig.’ Sw. discusses N.’s commentarii gram.,
his books de dis, de augur., de extis, de somnis, and
his astronomical writings.
No. 48. Reichert, Ueber den zweiten Theil der
Odyssee. A valuable investigation. ‘The writer of
υ 184 sqq. made use of the Telemachy and x, w. He
is the redactor of and responsible for the second half
of the Odyssey.’ R.’s treatment of x, y, and ¢ is
also very noteworthy.—Andreae Cricii Carmina ed.
Morawski. Very careful edition of C.’s (1477—1537)
poems.—Beitriige zur Geschichte des Humanismus in
Polen (Sb. Vienna Ac.). Is based on MS. letters, etc.,
and gives much new information.
No. 49. Franz, Mythologische Studien II. Dis-
cusses the ver sacrum and the sacrifice of kings. In
his attempt to reconstruct old customs from myths,
etc., F. goes frequently too far.—Demosthenis orationes
e rec. Dindorfii cur. Blass. ed. IV. vols. IH. and 111.
As in vol. I., the theory of responsion seems to play
too prominent a part in B.’s textual criticism. Kaz”
᾿Αριστογείτονος a’ and πρὸς Ζηνόθεμιν are considered
genuine.—Bellum Alexandrinum erkl. v. Schneider.
An excellent edition. Serves at the same time asa
critical ed., giving the most important MSS. readings.
The preface contains a refutation of Nipperdey’s view,
that Hirtius was the author of the B. Al.—List, Alt-
arisches jus gentium. The results can be regarded as
fundamental. L. is in strict opposition to Gump-
lowicz’s social theories and lays perhaps too much
stress on the plant-like growth of institutions and
social bodies.
LIST OF NEW BOOKS.
ENGLISH BOOKS.
Demosthenes’ Ovations against Philip, with Introduc-
tion and Notes by E. Abbott and P. E. Matheson.
Part I]. 4s. θά.
Goodwin (W. W.)
of the Greek Verb, re-written and enlarged.
pp. 492. Macmillan & Co. 14s.
Syntax of the Moods and Tenses
8vo.
THE CLASSICAL REVIEW.
Henry (Victor) A Short Comparative Grammar of
Greek and Latin for Schools and Colleges, trans-
lated from the second French edition by R. T.
Elliott. With an Introductory Note by Henry
Nettleship. Cr. 8vo. pp. xxx, 330. Sonnenschein
& Co. 7s. 6d.
Herodotus, The History of. Translated into English
by G. C. Macaulay. 2 vols., post 8vo. pp. 780.
Macmillan & Co. 18s.
Livy. Book IV. Edited by H. M. Stephenson.
12mo. Pitt Press. 2s. 6d.
—— Legends from Ancient Rome. Adapted and
edited, with Notes, Exercises, and Vocabularies,
by Herbert Wilkinson. 18mo. pp. 106. Mac-
millan & Co. 1s. 6d.
Moulton (R. G.) The Ancient Classical Drama: a
Study in Literary Evolution, intended for readers
in English and in the original. Cr. 8vo. pp. 476.
Clarendon Press. 8s. 6d.
143
Oman (C. W. C.) A history of Greece from the
Earliest Times to the Macedonian Conquest. With
Maps and Plans. pp. x, 535. Cr. 8vo. Rivingtons.
4s. 6d.
Simcoz (W. H.) The Language of the New Testa-
ment (the Theological Education Series). 12mo.
yp. 226. Hodder & Stoughton. 2s. 6d.
Stedman (A. M. M.) Greek Vocabularies for Repe-
tition, arranged according to subjects. 12mo.
pp. 70. Methuen. 1s. 6d.
Stone (E. D.) Selections from the Greek Tragedians.
12mo. pp. 202. Rivingtons. 3s. 6d.
Virgil. Aeneid, Book IV.: a translation by A. A.
Irwin Nesbett. Cr. 8vo. Sewed. Clive. 1s.
Xenophon, Works. Translated by H. G. Dakyns.
(4 vols.) Vol. I, Post 8vo. pp. 514. Macmillan
& Co. 105, 6d,
FOREIGN
Aeschylus. WOrestie. Traduction d’Alexis Pierron,
avec une préface par J. Lemaitre. Dessins de
Rochegrosse. 12mo. xx, 232 pp. Paris, Jouaust.
20 fres.
Barta (F.) Ueber die auf die Dichtkunst beziiglichen
Ausdriicke bei den romischen Dichtern. I. ‘dich-
ten’ und ‘Dichter.’ Programm. 8vo. 28 pp.
Linz.
Boutkowski-Glinka (A.) Petit Mionnet de Poche, ou
Répertoire pratique ἃ Usage des Numismatistes en
Voyage et Collectionneurs des Monnaies grecques,
avec Indication de leurs prix actuels et de leur
degré de rareté. 2 vols. 12mo. 418 pp. Berlin,
Kohler. Mk. 18.
Bidder (H.) De Strabonis studiis Homericis capita
selecta, Dissertatio inauguralis. 8vo. 56 pp.
Gedani. Mk. 1.50.
Rivista di Filologia e Vistruzione classica. Anno 18°.
Fase. 4-6. 8vo. 176 pp. Torino, Loescher. Per
annum, 121. 50.
Bouton (V.) Trois Lettres écrites en Octobre 1889 ἃ
M. le Comte de Werlé sur le Palatium de la Mon-
tagne de Rheims. Les Voies romaines qui y con-
duisoient. 4to. Paris, Bouton. 50 fres.
Caesar. De bello Gallico, illustrato da F. Romorino.
2° edizione rifatta. 8vo. 409 pp. Map. Torino,
Loescher. 31. 50.
Delbriick (H.) Die Strategie des Perikles, erliutert
durch die Strategie Friedrichs des Grossen. Mit
einer Anhang iiber Thucydides und Kleon. 8vo.
vii, 228 pp. Berlin, G. Reimer. 3 Mk.
Diels (H.) Sibyllinische Blatter. 8vo. 158 pp.
Berlin, G. Reimer. Mk. 2.80.
Di Giovanni (N.) La Topografia antica di Palermo.
S8vo. Palermo, Clausen. 35 lire.
Fehrnborg (O. J.) De verbis latinis in wo divisas
desinentibus. Diss. S8vo. 68 pp. Upsala.
Friedlander (.) Darstellungen aus der Sittenge-
schichte Roms in der Zeit von August bis zum
Ausgang der Antonine. 6t¢ neu bearbeitete und
vermehrte Auflage. 3. Theil. 8vo. xvi, 798 pp.
Leipzig, Hirzel. Mk. 14.
Fritze (K.) De Juli Frontini strategematon libro IV.
Dissertatio inauguralis. S8vo. 51 pp. Halle, 1888.
Mk, 1.20.
Fiihrer (G.) De particula ὡς eum participiis et prae-
positionibus junctae usu Thucydideo. Diss. S8vo.
38 pp. Miinster.
Geffcken (J.) De Stephano Byzantio. Commentatio
Hermanno Sauppio diem natalem octogesimum
BOOKS.
agenti a sodalitio philologorum Gottingensium
oblata. S8vo, 158 pp. Gottingen, Dieterich.
Mk. 2.
Guenther (P.) De ea, quae inter Timaeum et Lyco-
phronem intercedit, ratione. Diss. 8vo. 74 pp.
Lipsiae.
Handbuch der klassischen Altertums- Wissenschaft
herausgegeben von J. Miiller. Band II. 208 vollig
umgearbeite Auflage ; [griechische und lateinische
Sprachwissenschaft]. 2. Halfte. ὅνο. Miinchen,
Beck. Mk. 8.50.
Contents : Lateinische Grammatik bearbeitet
von Stolz und Schmalz. pp. 400-584.—Lexiko-
eraphie der griechischen und lateinischen Sprache
von Autenrieth und Heerdegen. pp. 585-635.—
Rhetorik der Griechen und Romer neubearbeitet
von Volkmann. pp 637~676.—Metrik der Griech-
en und Romer von Gleditsch. pp. 677-870.—
Indices xx and pp. 871-942.
Hart (G.) Uvsprung und Verbreitung der Pyramus
und Thisbe-Sage.. Dissertation. 8vo. 57 pp.
Passau. 1889. Mk. 1.60.
Hermes. Zeitschrift fir Classische Philologie heraus-
gegeben von G. Kaibel und C. Robert. Band
XXV. (in 4 parts). 8vo. (Heft 1, 160 pp.)
Berlin, Weidmann. Mk. 14.
Herodot’s Perserkriege. Griechischer Text mit er-
klirenden Anmerkungen. Fiir den Schulgebrauch
herausgegeben von V. Hintner. Theil 2. An-
merkungen. 2* verbesserte Auflage. 8vo. ili, 74
pp- Map. Wien, Holder. Mk. 1.20.
Hochart (P.) De V’Authenticité des Annales et des
Histoires de Tacite. Svo. Paris, Thorin. §8 fr.
Horling (W.) Sammlung lateinischer Satze. Kin
Beitrag zur Einiibung der grammaticalischen Re-
geln. Heft 1: Sitze zu den Regeln tiber den
Konjunktiv. 8vo. 94 pp. Paderborn, Junfer-
mann. Mk. 1.
Hoerschelmann (W.) Catull 68. 8vo. 24 pp.
Dorpat. 50 pfg.
Harzmann (F.) Quaestiones scaenicae. Diss. 8vo.
59 pp. Marburg.
Isocrates. Orazione per la pace commentata da C.
Tincani. 8vo. 199 pp. Torino, Loescher. 2 1. 50.
Judeich (W.) Persien und Aegypten im LV. Jahr-
hundert v. Christo. 8vo. 46 pp. Marburg.
Koch (W.) Paris vor Helena in der antiken Kunst.
Ein typengeschichtlicher Versuch. Diss. ὅνο.
72 pp. Marburg.
Kovroyovns, Σπυρίδων, Διορθώτικα εἰς τὰ ἔθνικα,
144
Στεφανοῦ τοῦ Βυζαντίου ἐν οἷς ἀκριβὴς περὶ τοῦ Σιμίας
λόγος. Ἑλληνικὸν ὀνομάστικον περιλαμβάνον τὸ
πρῶτον ἐξ ἐπιγραφῶν γνωριζόμενα ὀνόματα. Ἕκατον-
τὰς ά. 8vo. 97 pp. Erlangen. 6 Mk.
Larsen (S. C.) Studia critica in Plutarchi moralia,
Dissertatio inauguralis, 8vo. 152 pp. Kopen-
hagen, 1889. Mk. 3.
Lejay (P.) Inscriptions antiques de la Céte-d’Or.
8vo. Paris, Bouillon. 9 fr.
Lucretius. De naturarerum. Livre 1. Traduction
par H. Villard. 12mo. 23 pp. Paris, Jouaust.
Lyly (J. A.) Plootinos sielun substantsia alisundesta.
Diss. 8vo. 216 pp. Helsingfors.
Meingast (A.) Lateinische Stiliibungen. Programm.
8vo. 12 pp. Klagenfurt.
Mnemosyne. Bibliotheca philologica Batava, colle-
gerunt Naber, Leeuwen et Valeton. Nova Series.
Vol. XVIII. (in 4 parts). 4 partes. 8vo. (Heft 1,
112 pp.) Leiden. Mk. 9.
Miller (F.) Philipp 11. und die Athener in ihren
wechselseitigen Beziehungen zu einander. I. Pro-
gramm. 8vo. 21 pp. Olmiitz.
Miller (W.) Eine Terrakotta der Gottinger Samm-
lung. 8vo. 12 pp. Plate. Gottingen, Dieterich.
Mk. 1.20.
Museum, Rheinisches, fiir Philologie herausgegeben
von O. Ribbeck und F. Biicheler. Neue Folge.
Band 45 (in 4 parts). 8vo. (Heft 1, 160 pp.).
Frankfurt a/M., Sauerlinder. Mk. 14.
Niz (L. M. L.) Das 5. Buch der Cornica des Apol-
lonius von Perga in der arabischen Uebersetzung
des Thabit Ibn Corrah. Herausgegeben, ins
Deutsche iibertragen und mit eine EKinleitung ver-
sehen. 8vo. 48 pp. Leipzig, Hiersemann. Mk. 2.
Otto (P.) Quaestiones Strabonianae. Diss. 8vo.
60 pp. Leipzig.
Pascal (C.) Quaestiones Vergilianae ad eclogam IV.
spectantes. 8vo. 26 pp. Turin.
Peters (J.) De C. Valerii Flacci vita et carmine.
Diss. 8vo. 86 pp. Konigsberg.
Pichlmayr (Frz.) T. Flavius Domitianus. Ein Bei-
trag zur romischen Kaisergeschichte. 8vo. 100 pp.
Amberg, 1889. Mk. 1.80.
Plew (J.) Quellenuntersucbungen zur Geschichte des
Kaisers Hadrian, nebst einem Anhang iiber das
Monumentum Ancyranum und die kaiserlichen Au-
tobiographien. 8vo. iii, 121 pp. Strassburg,
Triibner. Mk. 5.
Plutarchi Moralia, recognovit G. N. Bernadakis.
Vol. If. 12mo. xxii, 557 pp. Leipzig, Teubner.
Mk. 3. .
Poiret (J.) Horace. Etude psychologique et lit-
téraire. 12mo. Paris, E. Thorin. 3 fr. 50.
Pollionis (C. Asini) De bello africo commentarius.
Recensuerunt, emandaverunt, adnotatione illustra-
verunt E. Wolfflin et A. Miodonski. Adjecta est
tabula photolithographica codicis Ashburnham-
ensis. 8vo. xi, 264 pp. Leipzig, Teubner.
Mk. 6.80.
Posnansky (H.) Nemesis und Adrasteia. Eine mytho-
logisch-archiiologische Abhandlung. ix, 184 pp.
Plate. 8vo. Breslau, Koebner. Mk. 7.
(Breslauer philologische Abhandlungen Band V.
Heft 2.)
Pusch (H.) Quaestiones Zenodoteae. (I.) Diss. 8vo.
30 pp. Halle.
Reichardt (C.) Landeskunde von Skythien nach
Herodot. Diss. 8vo. 134 pp. and 2 Maps.
Halle.
Rodemeyer (K. Τὸ Das Praesens historicum bei
Herodot und Thukydides. Diss. 8vo. 71 pp.
Basel.
THE CLASSICAL REVIEW.
Rohde (K.) Psyche. Seelencult und Unsterblich-
keitsglaube der Griechen. Hialfte I. 8vo. 294 pp.
Freiburg, Mohr. Mk. 7.
Rundischau, neue philologische. Herausgegeben von
C. Wagener und E. Ludwig. Jahrgang, 1890. (24
Nos.) 8vo. Gotha, F. A. Perthes. Mk. 12.
Schenck (L.) De Telesphoro deo. Dissertatio. 8vo.
55 pp. Gottingen. Vandenhoeck und Ruprecht.
Mk. 1.40.
Schmidt (O.) Der von den Romern (43-52) in Bri-
tannien gefiihrte Krieg. Programm. 8vo. 14 pp.
Wien.
Schuchhardt (C.) Schliemann’s Ausgrabungen in
Troja, Tiryns, Mykena, Orchomenos, Ithaka, im
Lichte der heutigen Wissenschaft dargestellt.
2 Portraits, 6 Maps, and 290 Illustrations. 8vo.
xii, 371 pp. Leipzig, Brockhaus. Mk. 8.
Schiitz (H.) Sophokleische Studien. Kritisch-exe-
getische Untersuchungen der schwierigeren Stellen
in den Tragédien des Sophokles. 8vo. viii, 450 pp.
Potsdam, Stein. 6 Mk.
Strabonis ἱστωρικῶν ὑπομνημάτων fragmenta collegit
et enarravit adjectis quaestionibus Strabonianis
Paulus Otto. 8vo. 350 pp. Leipzig, Hirzel.
8 Mk.
[Leipziger Studien zur classischen Philologie
Band XI. Supplement Heft. ]
Studien, Leipziger, zur classischen Philologie. Her-
ausgegeben von Ὁ. Ribbeck, H. Lipsius, C. Wachs-
muth. Band XII. Heft 1. 8vo. 281 pp. Leipzig,
Hirzel. Mk. 6.
Susemihl (Fr.) De Theoginae Orphicae forma anti-
quissima dissertatio. 8vo. xxi, pp. Greifswald.
Mk. 1.60.
Teuffel (W.S.) Geschichte der romischen Literatur.
Neu bearbeitet von L. Schwabe. δ Auflage.
Band I. ὅνο. xiv, 648 pp. Leipzig, Teubner.
Mk. 7.20.
Thiissing (J.) De temporum et modorum in enunti-
atis pendentibus apud C. Plinium Secundum usa.
Fasciculus 1. 8vo. 67 pp. Prag, Dominicus.
Mk. 1.20. :
[Prager philologische Studien. Heft 2.1
Toller (O.) De spectaculis, cenis, distributionibus
in municipils romanis occidentis imperatorum aetate
exhibitis. Dissertatio inauguralis. 8vo. 102 pp.
Altenburg. Mk. 2.
Treuenfeld. Der Zug der 10,000 Griechen bis zur
Ankunft am schwarzen Meer bei Trapezunt, darge-
stellt nach Xenophon’s Anabasis. 8vo. v, 146 pp.
3 Maps. Naumburg, Schirmer. Mk. 2.
Wegener (E.) De Aeschyli et Sophoclis fabulis ad
Herculem spectantibus. Diss. 8vo. 38 pp. Halle.
Welske (A.) Anmerkungen zur griechischen Syntax
mit Anschluss an Εἰ. Koch's griechische Schulgram-
matik. [Extract: Jahrbiicher fiir Philologie und
Padagogik.] 8vo. 24 pp. Halle, Buchhandlung
des Waisenhauses. 50 pfg.
Wentzel (G.) Ἐπικλήσεις Θεῶν sive de deorum cog-
nominibus per Grammaticorum Graecorum scripta
dispersis. Diss. 8vo. 156 pp. Gottingen.
Wochenschrift, Berliner philologische, herausgegeben
von C. Belger und O. Seyffert. Jahrgang 19 (52
Nos.). 1890. 4to. Berlin, Calvary & Co. Quar-
terly. Mk. 6.
fiir klassische Philologie, herausgegeben von
G. Andresen, F. Harder und H. Heller. Jahrgang
vii, 1890 (52 Nos.). 4to. Berlin, Gartner. Quar-
terly. Mk. 6.
Xenophon’s Anabasis.
ausgegeben von A. Weidner.
Leipzig, Freytag. Mk. 1.50.
Fiir den Schulgebrauch her-
12mo. xx, 266 pp.
The Classical Review
APRIL 1890.
THE GAME OF ‘HARPASTUM’ OR ‘ PHENINDA.’
So many learned writers, especially in
Germany, have treated of Greek and Roman
games that some apology may be needed for
opening the subject at all : but authoritative
as is their interpretation of most things,
the games at ball are precisely what we
have still left for doubt and conjecture.
Krause, Bec de Fouquiéres, Marquardt, and
even Grasberger in his admirable Lrzvehung
und Unterricht, while they supply a store-
house of references, do not seem to have con-
sidered enough the practical question—what
a player would be likely or able to do with
a ball, or what manner of rules could or
could not make a match between two sets of
players. Greek and Roman games are
loosely spoken of as somewhat like tennis,
or are even compared to golf!, although
there is no trace of any implement such asa
bat or racquet being used for any game bya
Greek or Roman until a late period—none,
as far as I know, earlier than the game
identical with polo which Cinnamus (vi. 5)
describes as played at Byzantium in the
reign of Manuel Comnenus. Strangest per-
haps asa failure to see essential differences in
games is the suggestion of Grasberger (op.
cit. p. 95) that harpastum may perhaps be the
same as this Byzantine game, though the
one is played on horseback with a long
curved stick, the other on foot with nothing
but the hand to propel the ball.
It will probably never be possible to lay
down with certainty all the rules of any
Greek and Roman game at ball, except those
of οὐρανία, which is simply a game of ‘ catch’ ;
Ὁ Such I conceive to be the meaning of Bec de Fou-
quitre’s statement (Jeux des Anciens, p. 203) that
the game of ἐπίσκυρος is ‘still played in Scotland.’
NO. XXXII. VOL. IV.
and of all the games harpastum is on the
whole the hardest to determine. But a com-
parison of ancient authorities will limit con-
siderably the field of discussion and will, I
believe, exclude many suggestions which have
been made,
The passages which form our authorities
for this game are Martial, iv. 19, vii. 32,
xiv. 48 ; Athenaeus, i. p. 15 ; Eustathius on
Od. 1x. 376; Pollux,.1.'32:; Sidenius, v. 17,
and especially the treatise of Galen περὶ τῆς
σμικρᾶς σφαίρας. On this last an elaborate
treatise has been written by Johann Mar-
quardt (Gustroviae 1879), whose authority
is accepted and quoted by Joachim Mar-
quardt in his Privatleben der Rimer. But I
cannot help thinking that Johann Mar-
quardt has started altogether on a wrong
path from supposing that Galen speaks of
three different games, and then trying by a
forced interpretation to fit im ἐπίσκυρος,
φενίνδα and ἁρπαστὸν as the three in ques-
tion. This idea may have originated in the
use of the plural by Galen (pp. 899, 900), τὰ
διὰ τῆς σμικρᾶς σφαίρας γυμνάσια, coupled with
the mention of ditterent degrees of exertion
suitable for different constitutions. In
reality however the wording of the treatise,
as well as its most natural interpretation,
should lead us to conclude that one game is
described. Its title is περὶ τοῦ διὰ τῆς σμικ-
pas σφαίρας γυμνασίου, and the plurals are
used in speaking of the different effects on
various parts of the body caused by different
phases of the game: eg. on page 902 he
says that he knows no other game so well
calculated to exercise all the limbs, either
severely or moderately as is requisite, τοῦτο
δὲ μόνον τὸ διὰ τῆς σμικρᾶς σφαίρας ὀξύτατον ἐν
L
140
μέρει καὶ βραδύτατον γενόμενον, σφοδρότατον
καὶ πρᾳότατον ὡς ἂν αὐτός TE βουληθῇς καὶ τὸ
σῶμα φαίνηται δεόμενον. Surely this passage
alone would exclude Marquardt’s interpreta-
tion of three different games suited for three
different ages or strengths. Have we never
heard in the modern game of football of a
man playing ‘goals’ because accident or
age has made him a less active runner than
he once was?! Briefly summarised Galen’s
argument in favour of the game is that it
not only exercises all parts of the body and
practises the eye, but also stimulates the
mind by a spirit of emulation. Of this last
he, as a physician, makes a great point, and
his remarks are valuable for our question as
showing that he is speaking of a real game
to be won or lost, and not of medico-gym-
nastics. He proceeds to prove that this
game suits all ages and constitutions, be-
cause each player can select that post or
duty in it which best suits his capacity : and
here again the forms of expression show that
he is deseribing one particular game, and
not three different games. The player may
take ὅσον ἐν αὐτῷ σφοδρότατον, or he may
choose the posts involving less exertion,
οὐδὲν γὰρ οὕτω πρᾷον, εἰ πράως αὐτῷ μεταχειρί
ζοιο : he may for instance take up a position
far from the centre, where he will have
chiefly to exercise his arms in throwing, or
he may have a great deal of running and
few long throws: or again he may take
that part which involves little rapid motion
but a great deal of grappling and wrest-
ling.
It will be seen that I have taken the
game which Galen calls that of the σμικρὰ
σφαῖρα to be harpastum. This can, I think,
be proved beyond a doubt. As the well-
known games at ball in which several
players are divided into two opposing sides
(defined as games κατὰ πλήθη or sphaero-
machiae) we gather only two from Eusta-
thius, Pollux and Athenaeus, namely ἐπίσκυ-
pos and ἁρπαστὸν or devivoa. Galen’s game
11 have mentioned football as a familiar instance
where players differing in activity and strength can
find suitable places in one and the same game ; but it
may be well to guard against the idea that football of
any sort was played in ancient Greece and Rome.
Johann Marquardt (among others) speaks of the ball
being kicked in harpasiwm as well as thrown, and
cites as his authority Bec de Fouquitres, who cer-
tainly makes the statement but cites no authority at
all. know of no passage in Greek or Latin litera-
ture which gives ground for this idea, which seems to
have arisen from the mention of jugglers, such as
Ursus Togatus (Orelli 2591), who caught and tossed
balls with their feet. Galen speaks of the exercise to
the arms in throwing: had kicking been allowed, he
would have mentioned that as exercising the legs, but
he assigns to them the exercise of running only.
THE CLASSICAL REVIEW.
is certainly not ἐπίσκυρος, the rules of which
are laid down by the first two writers with
remarkable precision, and, as we can hardly
suppose that the game which Galen selects
as the most complete and interesting would
be passed over by these writers, it is neces-
sary to identify 1t with harpastum: and
this Pollux does, when, speaking of devivéa,
he says εἰκάζοιτο δ᾽ ἂν εἶναι ἣ διὰ τοῦ μικροῦ
σφαιρίου ὃ ἐκ Tod ἁρπάζειν ὠνόμασται. It will
be seen also that what little can be gathered
from Martial about harpastum is in agree-
ment with this view: the ball with which
it was played must have been the smallest
and hardest of the four balls mentioned in
the Apophoreta, since the paganica (the ball
stuffed with feathers) is said to come
between the pila par excellence (i.e. the
trigon) and the follis as regards size and
hardness, and, as the follis was certainly the
largest, 1t follows that the harpastum was
the smallest. I have spoken above of the
pheninda as merely a synonym of harpastum.
It is difficult to understand how modern
writers can venture to treat these two as
separate games in face of the distinct state-
ment of Athenaeus, ‘70 δὲ καλούμενον διὰ τῆς
σφαίρας ἁρπαστὸν φαινίνδα ἐκαλεῖτο. As he
adds that it was his favourite game, the
flat contradiction of his statement in the
19th century seems all the more presump-
tuous ; and there is no conflict of authori-
ties to justify it ; for no ancient writer men-
tions them as distinct. Pollux alone in the
passage cited above goes so far as to say
that they might be ditferent, though he con-
jectures that they are the same. His note
of uncertainty might suggest that he was
more of a student than an athlete, but it
must be remembered also that the name
pheninda was, as Athenaeus tells us, gene-
rally superseded by the word harpastum,
though it was still retained in some places,
and is the only name appled to this game
by Eustathius and Clement of Alexandria.
The latter writer (Paed. 111. 10) in the words
σφαίρᾳ τῇ μικρᾷ παιζόντων τὴν φενίνδα affords
additional proof that pheninda and harpas-
tum were synonyms, if the foregoing remarks
upon Galen are correct : on the other hand,
if it is admitted that pheninda = harpastum,
the words of Clement will confirm the inter-
pretation of Galen. As to the correct spell-
ing of the word, devivda rather than φαινίνδα,
there can be little doubt that Meineke
(whom Marquardt follows) is right in
Hermes, 111. p. 455. Its connection in the
sense of misleading with φενακίζω (see Ltym.
Magn. s.v. φεννίς and Phot. Lex.) will be
understood from the description of the game.
THE CLASSICAL REVIEW.
That there was absolutely no alteration or
development in the rules of the game
between the time of Antiphanes and that of
Athenaeus—an intervalabout as long as from
the battle of Bannockburn to the present
day—is unlikely, but the main features of
the game must have remained and justified
the various names. The most characteristic
feature was the player who intercepted
(ἡρπαζε) the ball and who feigned a throw
(ἐφενάκιζε). This player is ὃ μεταξὺ of Galen
and the medicurrens of Sidonius, who is
obviously describing the same game as Galen.
To sum up the foregoing arguments:
Athenaeus asserts that pheninda is the same
as harpastum: Pollux thinks it is the game
with the μικρὰ σφαῖρα, which = ἅρπαστόν :
Clement of Alexandria says that it is played
with μικρὰ σφαῖρα. Taking these together it
appears clear that pheninda was the old
name for what was afterwards generally
called harpastum, and the older term still
lingered in some places when Clement wrote ;
and, further, that this game was so much
identified with the μικρὰ σφαῖρα that the
name of the ball expresses the game itself.
Tt must be admitted that a reconstruction
of the rules is in great measure guess-work,
but it seems to me that the following
account will explain and harmonize the
fragmentary descriptions in Greek and Latin
writers, and at the same time will not mili-
tate against common sense or the usual
habits of balls. The players were divided
into two sides, and each side had a base line,
for without this we cannot explain what
Galen says about στρατηγία and positions
won and lost. We must suppose then a
large rectangular ground with base lines at
each end, divided into two equal camps by a
line in the middle, which the ‘trames’ of
Sidonius must express.! So far the ground
resembles that of the ἐπίσκυρος, but the
resemblance seems to stop here. A special
feature of the game was, as has been said,
the ‘middle player,’ 6 μεταξὺ or medicurrens,
who is probably described by ‘vagus’ in
Martial vii. 22. One would indeed prefer
to imagine two middle players, so that each
side might have one, but the use of the sin-
gular in the authorities both Latin and
Greek seems to preclude this and to render
necessary some such explanation as is here
attempted. How the ‘innings’ of the medi-
currens terminated is not stated, but it may
be suggested that he gave up his place to
one of his opponents, whenever a point was
1 T strongly suspect that for ‘nee intercideret
tramitem nec caveret’ we should read οὐ intercideret
Χο,
147
scored against his own side. The main
object must have been to throw the ball so
that it should drop finally beyond the
enemy’s base line, thereby scoring a point ;
and we may suppose that it was started from
one or other base line and thrown from one
player to another, the opposite side thwart-
ing whenever they got an opportunity, and
throwing it back in the contrary direction.
The duty of the medicurrens was to catch it
as 10 went past (‘praetervolantem aut super-
jectam,’ Sidou.), which would give him a
better opportunity of throwing it over the
enemy's line, or into some unguarded spot
in their camp, where it might fall ‘dead’
and be started again, or of passing it on to
one of his own side who was advantageously
posted forward. Here would come in the
manceuvres from which the names of the
game arose: his intercepting the ball is ex-
pressed by ἁρπαστόν, the feint of throwing in
order to make his opponents rush ina wrong
direction suggested the name d¢evivda. One
among the essential points of difference
between this game and some others (e.g.
trigon) was that the ball might be taken at
the first bound as well as at the volley and
only dropped ‘dead’ when it fell a second
time, whereas at trigon the stroke was com-
plete as soon as the ball once touched the
ground (Petron. 27). This accounts for the
epithet pulverulenta (Mart. iv. 19), and the
alternative name of the ball arenaria, since
it was naturally more often on the ground :
hence also the expression ‘ rapit velox in pul-
vere’ (Mart. xiv. 48).
The duties of the other players may be
gathered from Galen and Sidonius. Some
of them (and naturally those who were less
active in running) stood near their own base
line, the ‘stantum locus,’ and only made
long throws towards the centre when they
got hold of the ball: others played nearer
the centre in what Sidonius calls the ‘area
pilae praetervolantis et superjectae,’ and ran
to whatever part of their camp the ball was or
seemed to be approaching, or ran forward,
so as to be ready to pass on the ball from
the medicurrens towards the enemy’s base :
in the event of the ball approaching their
own base there would be a rush back to the
rescue ; and this explains the words φυγή,
καταστροφή, catastropha, which we find in
Eustathius, Antiphanes and Sidonius. Lastly
some of the forward players, presumably the
strongest in muscle, were often engaged in
grappling with the medicurrens or with one
another in the endeavour to stop him from
catching and throwing the ball, or to pre-
vent his being stopped by others, as is
L 2
148
described by Galen “ ὅταν συνιστάμενοι πρὸς
ἀλλήλους καὶ ἀποκωλύοντες ὑφαρπάσαι τὸν
μεταξύ. . Hence the use of all the wrestling
terms, such as ἅμμα, ἀντίληψις, τραχηλισμός,
which may suggest some phases of the
‘Rugby game’: and this grappling by the
neck explains the otherwise obscure descrip-
tion of the harpastum-player in Martial
‘grandia qui vano colla labore facit,’ and
the line of Antiphanes ‘ οἴμοι κακόδαιμον τὸν
τράχηλον ὡς exw.’ Such an exposition of the
game will I think harmonize with the words
of Galen, which seem to me out of all har-
mony with the conceptions of recent writers
on the subject. ‘ You can,’ he says in effect,
‘ exercise a// your muscles, legs and arms and
chest, in throwing, running and wrestling,
and your eye at the same time in judging the
ball [{.6. if you are the medicurrens], or you
may take wrestling alone [as those who
thwart him], or ranning without much
throwing [as in the καταστροφή, and general-
ly in ‘forward’ play], or throwing without
much running [as those who play on the
line, the ‘stantes’].’ We can I think also
find here the explanation of the well-known
lines of Antiphanes, cited by Athenaeus (i.
p. 15) as descriptive of pheninda—
σφαῖραν λαβὼν
“ Ἀ ὃ ὃ \ 5, Ν δὴ 5, rE GP
τῷ μὲν διδοὺς ἔχαιρε, TOV ὃ ἐφευγ᾽ apa,
Ν δ᾽ > / \ τ 3 / ΄
τὸν δ᾽ ἐξέκρουσε, τὸν δ᾽ ἀνέστησεν πάλιν
κλαγκταῖσι [αἱ. πλαγκταῖσι] φωναῖς.
ἔξω, μακράν, παρ᾽ αὐτὸν ὑπὲρ αὐτὸν κάτω
ἄνω βραχεῖαν ἀπόδος, ἐγκαταστρέφου---
if we suppose the passage to describe part
only of the game, the action namely of the
medicurrens—having caught the ball he
throws it (δίδωσι) to A, one of his own side,
while he avoids B who tries to grapple with
him, and misleads (ἐκκρούει) C by a feint of
throwing it in some other direction, and
then, as the game goes on, he shouts again
to one of his own side to throw the ball,
high, low, &c., as may be needed to dodge
the opponent (παρ᾽ αὐτὸν ὑπὲρ αὐτόν), or
lastly to run back (ἐγκαταστρέφεσθαι) to
guard his own base. Or we may take the
interpretation of the last two lines (which
follow Meineke’s reading) to represent the
shouts of those opposing the medictrrens
and urging others to throw past him «ec.
The reading πλαγκταῖσι (which however does
not seem necessary) would imply a feint, like
ἐξέκρουσε.
The passage in Sidonius (17. v. 17) is not
only the best description of the game after
Galen, but also gives an amusing picture of
what may still sometimes be seen, an
THE CLASSICAL REVIEW.
elderly player in difficulties. “ Hie vir illus-
tris Philematius, ut est illud Mantuani poe-
tae, “ Ausus et ipse manu juvenum tentare
laborem,” sphaeristarum se turmalibus con-
stanter immiscuit : perbene enim hoc fecerat,
sed quum adhue essent anni minores.’—The
end of it is that, having first stationed him-
self ‘loco stantum’ (which I take to be the
line of back players, described by Galen as
only throwing ἐκ διαστήματος πολλοῦ, and
not running), he is next whirled by the
hurrying medicurrens into the middle area,
stumbles over the centre line (marked per-
haps, like the σκῦρος or λατύπη, with small
stones), is knocked down by a backward
rush of players (catastropha), picks himself
up at last and retires heated and out of
breath, which is bluntly expressed by
‘suspiriosus extis calescentibus.’ He is
more fortunate than the slave-boy in Dig.
9, 2, 52, § 4, who was knocked down in
much the same way, and broke his leg.
As to the passages from Antyllus (ap.
Oribas. i. p. 529) which complicate the ques-
tion in Bee de Fouquiéres, Marquardt and
others, it is to me perfectly clear that they
have nothing to do with this game, or any
other game properly so called; but describe
a course of medico-gymnastical exercises
wholly distinct from the contest between
sides which Galen gives us : in some of these
exercises the ball does not even leave the
hand but acts as a sort of dumb-bell in
extension motions. ;
It may be well to say a word in conclusion
about the argument at the end of Johann
Marquardt’s excursus as to three games,
drawn from the expressions ‘datatim, expul-
sim, raptim ludere.’ It is, I think, a pri-
mary cause of error in many writers (though
Joachim Marquardt in the main takes these
words rightly) that they have confused
methods of playing with games. 1. Datatim
ludere means simply to play by catching the
ball; throwing the ball for a catch being
dare, mittere or jactare, throwing it back
after a catch reddere, remittere : 2. expulsim
ludere on the contrary means to play by
striking the ball with the hand without hold-
ing it, the stroke used in our game of ‘fives’ ;
and the words eapellere, expulsare, repercutere,
ἀπόρραξις all apply to this stroke ; it could
be used equally by those playing together in
a game, or in solitary practice against a
floor or wall, as in Varro (ap. Non. 104, 27)
‘videbis in foro ante lanienas pueros pila
expulsim ludere’ : 3. raptim ludere describes
the play when the ball is intercepted by a
third person as it flies between two others.
These methods then are not games, but
THE CLASSICAL REVIEW.
strokes, which might be employed in various
games. In trigon, for instance, the play
might be either datatim or expulsim but not
149
raptim ; in harpastwm possibly all three, but
usually datatim and raptim.
G. E. Mar npin.
ON THE
SIGNIFICANCE OF SOME ECHOED PHRASES IN EURIPIDES’
HIPPOLYTUS.
Tue Hippolytus opens with the speech of
Aphrodite in which she reveals to the audi-
ence her grudge against the son of Theseus,
and her intended vengeance. He comes in
with his followers singing the praises of
Artemis: and the audience feels the irony of
the situation when they know and he does
not know what danger overhangs him. In
the remonstrances addressed to him by his
servant (107), τιμαῖσιν, ὦ παῖ, δαιμόνων χρῆσθαι
χρεών--ἂ door of escape seems to open. But
Hippolytus in his blindness rejects his
opportunity, and this in striking words (113),
τὴν σὴν δὲ Κύπριν πόλλ᾽ ἐγὼ χαίρειν λέγω. He
has formulated his sin, he has consciously
proclaimed that he adheres toit, and he leaves
the stage. Without doubt those last words
lingered in the mind of the audience, as the
summing-up of Hippolytus’ offence, and the
knell of his approaching doom.
But Hippolytus is not the only sinner in
the play nor the only object of divine
vengeance. Before the drama ends, Theseus
has played a part almost analogous to that
of his son.
Prima facie Theseus acts naturally in
believing the charges against Hippolytus
found in the hand of his dead wife. But
when without further inquiry he invokes on
his son a curse and banishes him from the
Jand, his action is over-hasty and insolent.
Again a way of escape is opened. In the
words of Hippolytus (1051-1055) :—
οὐδὲ μηνυτὴν χρόνον
δέξει καθ᾽ ἡμῶν, ἀλλά μ᾽ ἐξελᾶς χθονός ;
90) φΦ 3QX ΄ 50 Ν [2
οὐδ᾽ ὅρκον οὐδὲ πίστιν οὐδὲ μαντέων
φήμας ἐλέγξας, ἄκριτον ἐκβαλεῖς με γῆς ;
The gods have not left us without light.
Will you not use it? But as Hippolytus did
before, so Theseus now snaps his fingers at
the Divine power :—
ἡ δέλτος ἧδε κλῆρον οὐ δεδεγμένη
κατηγορεῖ σου πιστά: τοὺς δ᾽ ὑπὲρ κάρα
φοιτῶντας ὄρνεις πόλλ᾽ ἐγὼ χαίρειν λέγω.
These words contain and formulate the sin
of Theseus, as is shown clearly when Artemis
comes and denounces him,
ς ἃ 3
1321. ὃς οὔτε πίστιν οὔτε μαντέων Ore.
3, > Μ Lal
ἔμεινας, οὐκ ἤλεγξας, οὐ χρόνω μακρῷ
΄ὔ , 3 Ν an »” ?
σκέψιν παρέσχες, ἀλλὰ θᾶσσον ἡ o
ἐχρῆν
> Ἂν > ial Ν Ν ,
ἀρὰς ἐφῆκας παιδὶ καὶ κατέκτανες.
If I am right so far, 11. 113 sums up for
the poet and for the audience the sin of
Hippolytus,
τὴν σὴν δὲ Κύπριν πόλλ᾽ ἐγὼ χαίρειν λέγω,
and Hl. 1058-9,
Ν ΜΠ) ἊΝ la
τοὺς δ᾽ ὑπὲρ κάρα
lad 5», , ΣΟ ΝΣ / ,
φοιτῶντας ὄρνεις πόλλ᾽ ἐγὼ χαίρειν λέγω,
similarly sums up the sin of Theseus, can it
be doubted that Euripides purposely echoed
his own phrase in order to bring home to his
audience the recurrence of an old situation ;
can it be doubted that the audience recog-
nised the significance of the echo, and saw in
it what the poet intended they should see ?
It seems however to have escaped the
editors (Dindorf, Monk, Paley, Mahaffy,
Hadley) that we have here anything beyond
a mere verbal parallelism.
But there is another equally striking in-
stance in the same play where Euripides
seems to have again marked the similarity
of two situations by the use in each case of
the same phrase,
When the Nurse, under the pretext of
going for some drugs, is about to leave the
stage in order to acquaint Hippolytus with
Phaedra’s passion, Phaedra suspects her
intention and expresses the fear (520) μὴ μοί
τι Θησέως τῶνδε μηνύσῃς TOKW. The Nurse
answers ἔασον, ὦ παῖ, ταῦτ᾽ ἐγὼ θήσω
καλῶς. To the audience anticipating the
story, the irony of those words could hardly
fail to be striking. After an ‘aside’ only
three lines long, the Nurse leaves the stage
and at once works the irremediable mischief
which is the source of all the tragic events
that follow.
The secret has come out, and Phaedra and
150
the Nurse have to some extent exchanged
positions. Now it is Phaedra meditating an
act of tremendous consequences, an act
which shall wreck the happiness of Hippoly-
tus and of Theseus—the act of suicide—now
it is the Nurse who would fain avert the
evil.
705 ἀλλ᾽ ἔστι κἀκ τῶνδ᾽ ὥστε σωθῆναι, τέκνον.
Phaedra replies :—
cal ,ὔ Ν Ν Ν Ν > “
παῦσαι λέγουσα: καὶ τὰ πρὶν γὰρ οὐ καλῶς
παρήνεσάς μοι κἀπεχείρησας κακά.
> ? > Ν »” \ 4 /,
ἀλλ᾽ ἐκποδὼν ἄπελθε καὶ σαύτης πέρι
/ Ik > ‘\ ὃ Ν 5 Ν / r
φρόντιζ- ἐγὼ δὲτἀμὰ θήσομαι καλῶς.
The change to the middle is intelligible,
otherwise the phrase is the same, and I can-
not doubt that it was so chosen intentionally -
by the poet and that it touched a chord in
the hearts of the audience.
Yet I notice that Miss A. M. F. Robinson
in her translation of the play completely
drops even the verbal resemblance between
the two passages. The editors also are again
silent.
I have argued that the echoed phrase in
each case was not accidental, but was in-
tended by the poet and recognised by his
audience.
A further question arises:—Was_ the
repetition intentional not only on the part of
the poet, but also on the part of the dramatis
personae, of Theseus and of Phaedra? Are
we to consider them as consciously quoting
words used before, or as unconsciously
uttering what is nothing to them, but full of
significance to the spectators of the drama ?
We may at once dismiss the former supposi-
tion in the earlier of the two cases. Theseus
had not heard the words of Hippolytus which
THE CLASSICAL REVIEW.
he reproduced: his repetition of them is ob-
viously a case of tragic irony.
The other case is more doubtful.
Phaedra says (709)
W hen
ἐγὼ δὲ τἀμὰ θήσομαι καλῶς,
she has just been referring to that very con-
duct of the Nurse which followed the Nurse’s
words ταῦτ᾽ eyo θήσω καλῶς (521). More-
over these words had been spoken to herself.
It would seem very natural in her therefore
now to remember them and to apply them to
her own case. In a modern play we should
hardly hesitate to take this view of Phaedra’s
words.
But in a Greek play we must apply other
canons of interpretation, and when we re-
member to what an extent dramatic interest
was made by the Greeks to depend on Irony,
the contrast, that is, between knowledge on
the part of the spectators and blind uncon-
sciousness on the part of the dramatis per-
sonae, we may be inclined to consider Phae-
dra’s words in this light. It would follow
from such a conclusion that this case is ana-
logous with the former : that Phaedra no less
than Theseus unconsciously echoed words
used earlier in the play, but that this re-
petition was intended by the poet and recog-
nised by his audience as suggesting a parallel
in each case between two similar situations.
It seems to me clear that in both instances
the repetition is more than accidental. Tf
in the second instance we attribute to
Phaedra a cunscious use of the Nurse’s
words, the first instance must stand on its
own merits as a case of the character
described above. Can other instances of the
same kind be found in Euripides or the other
dramatists ἢ
G. C. M. Smiru.
MR. WHITELAW ON
AN interesting question is raised by Mr.
Whitelaw in the February number of the
Classical Review p. 12 as to this paragraph.
He supposes it to be the fragment of a lost
letter of St Paul to the Church of Corinth,
to which the Apostle alludes in 1 Cor. v. 9.
I wish briefly to inquire whether Mr. White-
law’s solution of the problem satisfies the
conditions ; and secondly whether there is a
problem to need solution.
(1) Against Mr. Whitelaw’s suggestion ob-
2 COR. vi. 11—vi. 1.
jection may be taken on the side of textual
criticism. There is no external evidence
against this paragraph. The probability that
copies of the Pauline letters were multiplied
from the very first (comp. Col. iv. 16 and the
case of the Ignatian letters, Zp. Polyc. xiii.)
and the remembrance of the critical questions
connected with Rom. xv. xvi. and Eph. 1. 1
make it difficult to suppose that the alleged
insertion could have been made without some
trace remaining in our extant authorities,
THE CLASSICAL REVIEW.
while the very perplexity of modern critics
forbids the assumption that early readers or
copyists placed the paragraph where it at
present stands. Again, internal evidence
does not tend to confirm the hypothesis.
There were in operation, it seems, in the
Corinthian Church two exactly opposite
tendencies. On the one side the new converts
showed a dangerous readiness to acquiesce
and share in heathen customs (comp. | Cor.
vii. 10 σὲ τὸν ἔχοντα γνῶσιν ἐν εἰδωλίῳ κατα-
κείμενον : χ. 20 οὐ θέλω δὲ ὑμᾶς κοινωνοὺς τῶν
δαιμονίων γίνεσθαι). On this matter in the
First Epistle the Apostle, uncertain of his
ground with the Corinthians, speaks with
caution and reserve, implying more than he
actually says. On the other hand there are
indications of a revolt, partial and temporary
no doubt, in the direction of an extreme
Puritanism (comp. 1 Cor. v. 9 ff. : vi. 1 ff.).
As to the former tendency the Apostle coun-
sels abstention from associations properly
heathen ; as to the latter he deprecates a with-
drawalfrom human society. Now in the pas-
sage in question, to whatever letter it belongs,
St. Paul is dealing with the former danger and
that in terms sufliciently explicit, e.g. τίς
κοινωνία φωτὶ πρὸς σκότος ; Tis δὲ συμφώνησις
Χριστοῦ πρὸς Βελίαρ, ἢ τίς μέρις πιστῷ μετὰ
ἀπίστου; τίς δὲ συγκατάθεσις ναῷ θεοῦ μετὰ
εἰδώλων; It is of course possible that the
Corinthians read into these words such a
definite misunderstanding of St. Paul’s
teaching as he alludes to in 1 Cor. v. 9; but
the gloss there implied seems to postulate
an Apostolic direction which would more
immediately suggest it. It is a natural, if
not a necessary, conclusion from St. Paul’s
language in 1 Cor. v. 9 ff. that he is there
referring to the exact phrase which he had
used in the lost letter—py συναναμίγνυσθε
TOpVOLs.
(2) But the more important question is
whether there is any problem here to need
solution. Is the position from which Mr.
Whitelaw starts—‘The passage, where it
stands, is without connexion before or
after ’—a tenable one? In answering this
question it is important to notice the charac-
ter of St. Paul’s second letter to Corinth.
Probably there is no literary work in which
the cross-currents of feeling are so violent
and so frequent. Again and again they
sweep the Apostle far away from his intended
course of thought and grammar. He
struggles back again, only to be once more
hurried away in yet another direction. Or,
to change the metaphor, we see a thought
bubbling up from the ground of the argu-
ment fresh and vigorous. But at once it
151
passes beneath the sudden rising-ground of
some new idea; at length it appears again
tinged with the soil through which it has
flowed. Thus ii. 17 πρὸς ταῦτα τίς ἱκανός ;—
but the question is at once buried beneath
the thought of the contrast between St. Paul
and others ; when it emerges once more in
iii. 5 it is in an altered and an apologetic
form. Again, in iv. 1 St. Paul writes οὐκ
ἐγκακοῦμεν GAka—we expect ἀνδριζόμεθα or
θαρροῦμεν to express the proper contrast.
But there flashes into St. Paul's mind the re-
membrance of eruelaccusations. So breaking
through the strict sequence of thought, he
continues ἀπειπάμεθα τὰ κρυπτὰ τῆς αἰσχύνης.
The original idea and phrase reappear in iv.
16 οὐκ ἐγκακοῦμεν GAAG...... ; but the inter-
vening context (¢.g. 6 θάνατος ἐν ἡμῖν ἐνεργεῖ-
rat) influences the words which follow,
‘though our outward man is decaying, yet
our inward man is renewed day by day.’
But here again the Apostle is carried away
by the thought of the contrast between the
weakness of the present and the glory of the
future. Once more the contrast to ἐγκακοῦ-
μὲν appears in the θαρροῦντες of v. 6 and,
the grammar meantime having suffered ship-
wreck, in the θαρροῦμεν of v. 8 and the φιλο-
τιμούμεθα of v. 9.
Remembering this characteristic of the
Epistle we turn to vi. 11—14....1) καρδία ἡμῶν
πεπλάτυνται (cf. Ps. exvili. 32 ὅταν ἐπλάτυνας
τὴν καρδίαν μου)... πλατύνθητε Kat ὑμεῖς. At
this point the Apostle recalls the condition
of the Corinthian Church and their danger
of a false tAatvocpds. He must emphatically
and abruptly guard his words: μὴ γίνεσθε
ἑτεροζυγοῦντες ἀπίστοις (comp. Lev. xix. 19
τὰ κτήνη σου οὐ κατοχεύσεις ἑτεροζύγῳ). That
this is the connexion of ideas becomes, I ven-
ture to think, almost certain when we com-
pare another passage of the Law: πρόσεχε
σεαυτῷ μὴ πλατυνθῇ ἡ καρδία Tov καὶ παραβῆτε
καὶ λατρεύσητε θεοῖς ἑτέροις καὶ προσκυνήσητε
αὐτοῖς (Deut. xi. 16).
But what of the connexion of the para-
graph in question with the succeeding con-
text? The Apostle resumes no doubt his
earnest pleading for sympathy. But a
shadow has passed over his enthusiasm.
Above he had written, ὡς τέκνοις λέγω. Now
he writes, οὐδένα ἠδικήσαμεν, οὐδένα ἐφθείραμεν,
οὐδένα ἐπλεονεκτήσαμεν. The language he
had used just above (v. 1) καθαρίσωμεν ἑαυτοὺς
ἀπὸ παντὸς μολυσμοῦ σαρκὸς καὶ πνεύματος,
calls to his mind the charges made by some
of the Corinthian partisans against himself
—-the charge of corrupt motives and the
charge, which seems to be lurking in the
background of some passages, of antinomian-
152
ism (iv. 2 Comp. 1
Thess. ii. 3 ἡ yap παράκλησις ἡμῶν οὐκ ἐκ
πλάνης οὐδὲ ἐξ ἀκαθαρσίας οὐδὲ ἐν δόλῳ). Thus
the connexion of thought at the beginning
and at the end of the paragraph is clear and
natural.
If the character of the Epistle be remem-
“ ΄
τὰ κρυπτὰ τῆς αἰσχύνης.
THE CLASSICAL REVIEW.
bered, its deep emotion and its consequent
abruptness of thought and expression, the
reasons for believing that the paragraph is
part of the original letter seem to me valid
and convincing.
F. H. CuHase.
STUDIES IN
THe majority of classical students are
wont to underestimate or neglect this writer.
His ‘Gallic War’ to the average student
has been chiefly a Latin primer and a collec-
tion of examples to illustrate oratio obliqua
or the various functions of the ablative
absolute. With a feeling akin to condescen-
sion the mature student—unless he himself
be a teacher of boys—will take up this
author asa man would take up a pair of
dumb-bells which once had done him service.
The present writer distinctly recalls the
earnest admonition on this matter uttered
many years ago (1873) by one who is a
master in this department, Prof. Emil
Hiibner of Berlin. This distinguished
Latinist urged the professional classical
students to return to Caesar with their
matured powers, if they wished to do him
justice.
In Germany and France indeed, more than
in England and in the United States, very
many distinguished scholars have devoted
themselves to this writer. It might indeed
appear presumptuous to attempt anything
in this field after Druwmann, Merivale, and
Mommsen, or even to supplement the work
of Schneider (whose acute observation has
furnished a great deal of Long’s edition),
Nipperdey, Riistow, Herzog, Kraner, Dober-
enz, Dinter, Kichly, and Holder, whose
edition of 1882, while probably not displacing
the fearfully expensive edition of Nipperdey,
is distinguished by its index verborum, and
thereby probably suggested the preparation
of the special lexica by Merguet, Meusel, etc.,
all of which began to appear after 1882. Of
those who combined the study of the subject-
matter with that of the text, no one deserves
a higher place than H. 7. Heller of Berlin.
For his exhaustive analysis of Caesar’s syn-
tax and the application of the results gained
to the tasks of writing Latin, etc., Hey-
nacher deserves more than passing notice.
The topographical and antiquarian studies
of Napoleon III. and Goeler have been
CAESAR 1.
worthily continued by v. Kampen of Gotha,
whose special wall-map of Gaul deserves a
place in every Latin school-room. Apart
from various notices in Polybius, Caesar is
the first of national historiographers to the
consciousness of educated Frenchmen, and
eminently so in the eyes of French anti-
quarians.
A long line of investigations in the Revue
Archéologique, particularly within the last
twenty-five years, attest the earnest devo-
tion of French students, by whom the coins,
weapons, inscriptions, burial-places, social
customs and ethnology of Gaul have been
accurately studied. Pre-eminent among
these scholars is Alexandre Bertrand, the inde-
fatigable curator of the Musée de St. Germain.
Diefenbach’s Celtica and Lindenschmit’s
(Mainz) Studies of Roman, Gallic and
German arms, etc., must also be mentioned
here.
ΤΕ
In the case οἵ Caesar’s book (called com-
mentarii with a simplicity which was pro-
ably designed) the why and the when would
seem to the student to be closely connected.
For the benefit of those who may wish to
take up the subject, I name a few of the
most competent writers dealing with design
and time of composition. H. Kéchly, Caesar
und die Gallier, Berlin 1871, pp. 31-32;
Kramer, Praecfatio of his Tauchnitz edition,
1861, pp. xvii. sqg.; Mommsen, R. G. in.
(ed. 6), pp. 615 sgg. (Students in search
of a complete bibliography had __ better
turn to ‘Teuffel, or Prof. MHiibner’s
Grundriss d. Rom. Literaturgeschichte,
or to a special treatise by Major Max
Jiihn’s ‘Caesar's Commentarien und ihre
literarische und kriegswissenschaftliche
Fortwirkung, Berlin, 1883: Ernst Sieg-
fried Mittler und Schn.) Mommsen presents
his view in the categorical and compact way
peculiar to himself. Merivale does not seem
to have attempted to solve the particular
THE CLASSICAL REVIEW.
question as to the occasion and design of
the composition of the Commentaries.
In looking for the motives of Caesar’s
book, we may at once reject the theory of
mere literary ambition or of the pure plea-
sure of composition—motives such as did
animate a Cicero, whose restless and facile
pen was ever seeking new fields, and whose
semi-morbid consciousness was ever on the
alert as to the figure he was cutting in the
world. Nor can Xenophon be compared when
—probably in the calm repose of his squire’s
life at Skillus—he set down the narrative
of events quorum et ipse pars magna fuit,
and indulged in his peculiar hero-worship.
Caesar was not vain, apart from sensitive-
ness about his baldness. He was too pene-
trating for such a weakness, although he
was no doubt conscious of the permanence
which his work was destined to enjoy: e.g.
when inserting the account of the grim
valour of some Gallic patriots, as in vil. 25,
1, ‘ Accidit suspectantibus nobis quod dig-
num memoria visum praetereundum non
existimavimus (compare on the other hand
Bell. Civ. 111. 17, 1, Quibus rebus neque
tum ‘Caesar respondendum existimavit neque
nune ut memoriae prodantur, satis causae
putamus). Caesar was eminently versatile.
A political manipulator on an enormous
scale, he became a great captain after his
fortieth year from political necessity. But
during the seven years from 58-51 (apart
from short periods of enforced leisure, in one
of which, viz. when returning! to Gaul from
his cisalpine province, Suet. 56, he is said
to have composed his treatise De Analogia),
his enormously varied occupations would
have prevented him from gratifying mere
literary ambition. His correspondence with
his political henchmen in or near the capital,
variously but palpably enough reflected and
suggested in Cicero’s letters, must, in itself
alone, have been tremendous.
Let us now turn to the passage which may
to-day be considered as the accepted starting-
point for the whole question, Hirtius, B. G. 8,
1: ‘Ceteri enim quam bene et emendate, nos
etiam quam facile atque celeriter eos perfec-
erit scimus.’ This is generally taken as
meaning that Caesar wrote the entire seven
books in one uninterrupted effort (Nipperdey,
praef. 2 sqq.). And indeed Caesar had
strong motives at that particular time (51 B.c.)
to set himself right in public opinion, Faith-
ful as his political agents at Rome were—
and he had spent millions in securing some
τ Cf. the practice of Cicero in his Cilician procon-
sulate, ad. Att. v. 17, 1: ‘hane epistolam dictavi
sedens in raeda, cum in castra proficiscerer.’
153
of them, e.g. Curio—the senatorial party in 51
and 50 was striving, with 8411 the machinery
of the home government except the tribunate
and the comitia at its command, to reduce
Caesar to the ranks of a private citizen.”
Α fewyearsearlier, before the deathof Julia
and the estrangement from Pompey, the
machinery of the government had run
smoothly enough for Caesar’s interest. This
status is well shown in Cicero’s speech of 56
B.c. de Provinciis Consularibus, when Caesar’s
imperium was extended by five years. But
still, if prominent members of the sena-
torial party as early as 58 B.c. intimated to
Ariovistus, the German invader of Alsace,
that they would be delighted if he were to
kill Caesar (B. G. 1. 44), we may well a
Jortiort infer that in 51, when the desperate
hatred of the Optimates against Caesar had
reached a much higher pitch, when the
triumvirate had been disrupted—Crassus
dead at Carrhae, Pompey flattered back to
the leadership of the Optimates—we may
infer, I say, that at this stage Caesar’s acts
of his seven years’ imperium were in the
ruling circles at Rome not only not esti-
mated fairly, but distorted, interpreted un-
favourably, and in every way represented to
his discredit. This view of partisan con-
demnation, however met by partisan com-
mendation on the other side, found vent not
only in motions such as the familiar one of
Cato’s, to surrender Caesar to the survivors
of the foully and treacherously massacred
Usipetes and Tencteri(55 B.c.), but must have
also been reflected in the political pamphlets
of the day. The present writer shares with
other students of this subject the impression
that the general judgment of Suetonius
about Caesar’s proconsulate was drawn from
suchsources (rather thanfrom Asinius Pollio,
as some have surmised). There are it is true
judgments im Suetonius, though probably
not ef Suetonius, which reflect a /ater stand-
point, e.g. Div. Tul. 22, ‘ex omni provinci-
arium copia Gallias potissimum elegit cuius
emolumento et opportunitate sit materia
triumphorum.’ In the same chapter we
read also: ‘et initio quidem Galliam Cisalp-
inam Illyrico adjecto lege Vatinia accepit,
mox per senatum Comatam quoque, veritis
patribus ne si ipsi negassent, populus et
hance daret. Quo gaudio elatus non temper-
avit quin paucos post dies frequenti curia
tactaret invitis et gementibus adversariis
adeptum se quae concupisset, proinde ex eo
insultaturum omnium capitibus,’ with further
2 Cf. particularly the reports which Caelius sent to
Cicero in Cilicia in 51 B.c., Cicero ad. Fam. viii.
8, 9.
154
unsavoury allusion to the intrigue with Nico-
medes.
This in the senate—before the cat-and-dog
performances of Cato and Clodius and the
Milonian era had done so much to lower the
gravitas of that body. The passage last
quoted (or its source) smacks of vindictive
partisanship. The narrative of Suetonius
continues in the same spirit (24) : ‘ Nec deinde
ulla belli occasione ne iniusti quidem ac
periculost abstinuit, tam foederutis quam
infestis ac feris gentibus wltro lacessitis, adeo
ut senatus quondam legatos ad explorandum
statum Galliarum mittendos decreverit ac
nonnulli (Cato and his followers) dedendum
eum hostibus censuerint.’ The animus here
is palpable enough. To this we may add
(ib. 54), ‘in Gallia fana templaque deum
donis referta expilavit, urbes diruit saepius
ob praedam quam ob deliccum.’
Let us now turn to Caesar’s own writing
about his proconsulate, taking up a number
of passages in which, particularly if they be
taken together, the apologetic and semi-con-
troversial character of the commentaries
becomes probable, if not evident,
The Helvetii, having failed in their effort
to pass the Rhone below Geneva and thus
enter the province, had passed through the
defiles of the Jura, crossed the Saone, with
heavy losses, not much to the north of
Lyon, and were marching away from the
provincia, But their ultimate destination
was Saintonge, north of the Gironde, and the
THE CLASSICAL REVIEW.
success of their attempt, says Caesar, would
threaten the provincia of which they would
become ‘neighbours’ (Santonum fines... .
qui non longe a Tolosatium finibus absunt,
quae civitas est in provincia I. 10, 1). Non
longe indeed !—about 200 mules, though
Caesar is right in noting the absence of all
natural barriers.
Another motive for his Helvetian cam-
paign was the earnest petitions! of the
Aedui to save them from the Helvetian
locusts (I. 11, 3-4). These things, he inti-
mates, and not his ownambition determined
him to pursue the Helvetii (I. 11, 6). The
insolence and arrogance of the Helvetian
chief Divico, who represented the memory of
a Roman humiliation, is cleverly introduced
by Caesar in such a way that the latter
appears as the avenger of national honour and
restorer of Roman prestige. Caesar’s solemn
reference, in the Herodotean vein, to the
custom of the immortal gods to be indulgent
to overweening pride with designed patience
in order to make the ultimate catastrophe the
more telling—all this is a trifle forced,
although no doubt flattering to Roman
pride.
E. G. Srmter, Pu. D.
New York.
1 As to their intensity and sincerity we have but
Caesar’s statement. It is curious that the movement
of the Helvetii should have had the active co-opera-
tion and sympathy of that party among the Aedui
who followed the leadership of the most powerful ‘of
their barons, Dumnorix.
(To be continued.)
HORATIANA.,
In venturing a few emendations of the text
of Horace, I do not think it worth while to
support them by any express appeal to pa-
laeography. Such an appeal implies a know-
ledge, or at least a theory, of epochs in the
tradition, whereas we can fix no dates in
the ancestry of our Horatian MSS. All our
oldest copies are written in minuscules: all
come from Northern Europe: all are about
900 years later than Horace’s time and
nearly all are incomplete. Again, at almost
every ditliculty, we find either that our MSS.
are agreed or that they vary, in a haphazard
way, between (substantially) only two read-
ings, both of which usually make some sense.
The facts in general suggest an incorrect
and blurred archetype and two restorations
of it: but the details! suggest further that
there was a separate archetype of the Lyrics,
of the Satires and of the Epistles. Our
MSS. seem to be composite in two ways:
firstly, the separate works were taken from
distinct sources ; and, secondly, in each work
the scribe had variants before him, between
which he selected. The date of these vari-
ants is the difficulty, for those which seem
to be late (such as pardus for pagus in C.
iii. 18. 12, or comptus for coronatus in C. 1].
7. 7) are just as well supported as those
1 For instance with one exception, res sponsore for
responsore in Ep. i 16. 43, all the striking readings
of the oldest Blandinian are in the Satires.
THE CLASSICAL REVIEW. 15
which seem to be early. Where the history
of a text isso vague, it is useless to be talking
of capitals and uncials and half-uncials and
the like or of peculiarities in the writing of
any particular MS. : an emendation is plau-
sible only if it improves the sense and sug-
gests an error possible in almost any style of
writing.
Carm. ui. 5. 84—37
Μὲ Marte Poenos proteret altero
qui lora restrictis lacertis
sensit iners timuitque mortem
hic unde uitam sumeret inscius (aptius)
pacem duello miscuit. O pudor! ete.
The MSS. all have λὲς in 37, and most have
inscius, but several aptius. Bentley, object-
ing to an abrupt stop at mortem (cf. iv. 9.
50) and also thinking that hic should rather
be dle (as in 1. 32), proposed ‘ timuitque mor-
tem hinc u. u. 5. aptius pacem et duello m.’
The reading of Haupt and many others
‘hic—anxius’ does not touch Bentley’s objec-
tions. The only other suggestion is Mr.
Housman’s ‘fine unde uitam swmere tustius
pacemque bello miscuit.’ The sense which
Bentley and Housman assign to hine, viz. ‘a
pugna, ferro, manu,’ seems to me too vague
for the occasion. The sword from which
the soldier should fear death is the enemy’s:
the sword with which he should win life is
his own. I would propose, therefore, ‘tim-
uitque mortem inc unde uitam sumere twr-
pius. Hine unde means ‘from the enemy
from whom’ it is disgraceful to take life
(cf. Carm. 111. 11. 38, unde non times: Sat.
i. 4. 6, hine omnis pendet). The soldier is
convicted of a real confusion of thought,
which is put epigrammatically in ‘pacem
duello miscuit.’ In peace, as Ruskin insists,
it is brave to save your life: in war it is
brave to lose it. For miscere in the sense
of ‘to make no difference between’ cf. Zp.
1. 16. 54 miscebis sacra profanis.
Carm. iii. 14. 1O—12
uos, o pueri et puellae,
iam uirum expertae, male ominatis
parcite uerbis.
In 11, three good MSS. have eaperte: the
scholl. and many MSS. have male ominatis,
but most of the latter have male nominatis,
an unparalleled use. I propose
‘jam uirum eapectate. male ominatis’ etc.,
where wirwm means, as often, ‘the pattern
man’ and parcite is addressed to the whole
throng of sightseers. The abruptness is
similar to ‘fauete linguis’ in Οὐ. iii. 1. 2
and elsewhere. The rhythm is in Horace’s
later style, but the ode is late (B.c. 24), like
i, 12, which begins ‘Quem uirum aut
? 5
heroa.’
Carm. iv. 2. 49
teque dum procedis io triumphe ! etc.
All MSS. and scholiasts have teque, a
famous crux. I propose ‘ zoque,’ io being a
monosyllable as in Catullus 1]xi. on which
see a long note of Munro’s in Critic. and
Eluc. pp. 1386—138.
Sat. 1. 1. 108, 109
ailluc unde abit redeo nemon’ ut auarus
se probet ac potius laudet diuersa sequentes.
All existing MSS. have nemon’ ut or ne
non ut: the Blandinius antiquissimus alone
had qui nemo ut auarus, which can be ex-
plained only by Mr. Palmer’s desperate ex-
pedient of supposing an ellipse of ‘ fiat’ (se.
qui fiat ut nemo auarus etc.). I would sug-
gest
‘nemone suam rem (sua ré)
Sed probet’ ete.
Orelli, Holder and others read ‘nemo ut
auarus, but the exclamation, rather than
the question, seems to me in place at the end
of a satire which shows the folly of discon-
tent, and I feel to want ‘sed’ or nothing.
For ‘ probo’ and ‘laudo’ used together L.
and §. give several examples under the
former word. A somewhat similar corrup-
tion appears in Fp. 11. 1. 226, where for
60 rem many MSS. have tem fore.
Yat. τι. 2. 9—14.
leporum sectatus equoue
lassus ab indomito, uel si Romana fatigat
militia assuetum graecari, seu pila uelox
molliter austerum studio fallente laborem,
[sew te discus agit, pete cedentem aera disco, |
cum labor extuderit, ete.
‘Totus hic locus uidetur claudicare’ says
Lambinus, and all other editors agree. The
remedy is to omit ]. 13 : then Romana militia
and pila uelox are alternative modes of tiring
yourself, and fatigat goes with both. The
unusual form si—seu, which is due to the
preceding wel, caused the interpolation.
Sat. ii. 3. 208, 209
qui species alias uert scelerisque tumultu
permixtas capiet, commotus habebitur ete.
Holder’s note is ‘wert sceleris ayvEDLv
ueris celeris Ἐ(φψ)ὴλ (Al) weris sceleris g.’
Porphyrio seems to have had weris celert. I
would propose
‘alias weris cerebrique tumultu
permixtas,’
For alius with the abl. of comparison cf. Hp.
i. 16. 20 (aliwm sapiente). Cerebrum in
150
the same sense occurs in]. 75 of this satire ;
tumultus mentis in Carm. 11. 16. 10: permix-
tus in the sense of ‘interchanged’ isin Lucr.
111. 749. Cf. also miscere above.
Ep. i, 2. 29—31 (Aicinoi iuuentus)
Cui pulchrum fuit in medios dormire dies et
ad strepitum citharae cessatum ducere curam.
All MSS. have cessatwm, and most have
curam, but several (incl. Bland. antiq.) have
somnum. This seems a clear case of damage
to the archetype and a heroic remedy is
therefore not inadmissible. I think the
original reading may have been
cessare et ludere et ungui.
THE CLASSICAL REVIEW.
The same words occur again in Lp. i. 2.
183, and the same jingle (cessare et quaerere et
uti) in Ep. i. 7. 57. It seems in Horace’s
manner to repeat these jingles, as libertino
patre natum in Sat. 1. 6, 6, 45, 46 and some-
thing like it in 21: also Jaeuo suspensi ete.
in Sat. i. 6. 74 and Lp. i. 1. 56, where Orelli
gives five more examples of repeated lines.
Here in particular the string ‘cessare et
ludere et ungui’ is the more appropriate
because Horace is thinking of Homer’s
αἰεὶ δ᾽ ἡμῖν δαίς τε φίλη κίθαρίς τε χοροί τε
“ , 2 ore Ν ia Ν ᾿ > z
εἵματά τ᾽ ἐξημοιβὰ λοετρώ τε θερμὰ καὶ ευναί.
in 6. 248, 249. J. Gow.
THE ORIGIN OF ‘OMENTUM™.’
‘Foups passing from one viscus to the
other...are called omenta from their his-
torical connection with extispicious augury,’
says Professor Macalister in his great work
on Anatomy just published (p. 391).
Modern philologists have passed over such
a derivation with the silence of contempt,
but seeing from the statement quoted above
that this view is not, as might have been
thought, altogether defunct, [ have under-
taken the present article in the hope of
showing that, however dubious the origin of
Omentum may be, the word has no connec-
tion with omen.
Bochart in his Hierozoicon (1796) hesit-
atingly mentions the derivation from omen,
as the view of others, not his own: ‘ Ex
omenti statu et situ haruspices multa coniec-
tabant, proinde Lucanus ubi de inauspicato
Aruntis sacrificio, lib. I. vers. 624,‘‘ Cor iacet,
et saniem per hiantes viscera rimas Emittunt
produntque suas omenta latebras.” Et in
Oedipo Senecae Manto extis inspectis Tire-
siae patri tam caeco inter caetera dicit,—
“non molli ambitu Omenta pingues viscerum
obtendunt sinus.” Itaque Omentum ab
omine quidam deducunt. Quod etsi prae-
stare nolim, tamen puto verisimilius quam
quod plerique statuunt, Omentum ab omaso
dici ; quia in omaso prima est brevis’ (Lib.
111.). Even omasum itself is credited with
the same origin by Joannes Janua (Du
Gange, Glossarium, Vol. V1I.), ‘Omasus (sic) i.
tripa vel ventriculus qui continet alia viscera
quia in ipso rerum eventus inspiciebant.’
Now, if Omentum be derived from omen,
it would surely be because the Omentum was
the part par excellence, from which omens
were drawn.
But so far from this being the case, it is
difficult to prove that the omenta were ever
the source of omens, and it is practically
certain that, if ever so used, they were
merely a minor source.
Tt will be observed that in the two pas-
sages quoted by Bochart there is no mention
of omens being taken from the omenta ; in
both, they are simply named as parts seen
when the viscera are exposed.
If references to the extispicium be com-
pared, it will be seen that the liver and the
nobler organs were .themselves the chief
objects of inspection, e.g. ‘Nam non placet
Stoicis, singulis iecorum fissis aut avium
cantibus interesse Deum’ (Cie de Div. 1. 52) :
and again (ibid.), describing an ominous
victim, ‘in extis bovis opimi cor non fuit’:
and again, ‘ postero die caput in iecore non
fuit’ : cf. also ‘ Quid enim habet haruspex cur
pulmo incisus etiam in bovis extis dirimat
tempus?’ (Cic. de Div. I. 39), and similar
passages in Plin. H.W. XI. 71 and Val. Max.
I. 6, 13, ete.
Later writers on the subject have noted
this order of importance, e.g. Caspar Peucer
(1593) in a passage too long to quote, but I
give the marginal headings, (1) ‘victimis
explorandis cur epar primo consideratur’ :
(2) “ Cordis consideratio in extispicina’ : (3)
‘ Viscerum minus principalium consideratio’ :
the last section beginning thus, ‘Corde
explorato ad fellis, lienis, pulmonum, et mem-
branarum viscera ambientium consideration-
em se vertebant’ (Comment. de praecipuis
Divinationum generibus, p. 361), and M.
Johannes Praetorius (1677) in a pamphlet
published with others on kindred subjects,
‘ Aruspicium, cum ex inspectis avium aut
THE CLASSICAL REVIEW.
aliorum animalium sacrificatorum extis prae-
cipue iecore (unde ἡπατοσκοπία dicta) colli-
guntur futura.’
The very name extispex occurring at least
as early as Cicero, and the use of ‘exta’ by
him, not ‘ viscera,’ would seem to show that
it was the nobler organs (exta), not the
whole entrails (viscera), from which divina-
tions were chiefly made. The word Omentum
was certainly in use before Cicero’s time
(Catullus B.c. 50), yet in the whole of the
De Divinatione, though referring frequently
to the extispicium, he never once uses the
word. Is it possible, if the Omentum were
an important source of omens, as it surely
must have been, if derived from omen, that
Cicero in such a treatise should never once
mention it ?
Further, if Omentum played such an im-
portant part, we should expect to find the
equivalent ἐπίπλοον used in Greek divination,
but I cannot find a single passage in which
it is so used; the word is σπλάγχνα with the
same distinction as in the Latin, σπλάγχνα
(the nobler organs), not ἔντερα or κοιλία (the
whole entrails).
Again, if the derivation from omen be
correct we should expect some traces of the
connection in the earliest writers using the
word. It occurs as early as Catullus, ‘ Natus
ut accepto veneretur carmine divos, Oment-
um in flamma pingue liquefaciens,’ then in
Celsus Lib. IV. cap. 1, al., who, though he is
usually careful to explain the origin of
anatomical terms, has nothing to say on
Omentum. Other passages in chronological
order are, Pers. VI. 74 ‘Ast illi tremat
omento popa venter,’ id. 17. 47 ‘Tot 0101 cum
in flammis iunicum omenta liquescant’ :
Plin. Δ΄. ΗΠ. XI. 37, 80 ‘Ventriculus atque
intestina pingui ac tenui omento integuntur’
(it is noteworthy that just above he has been
speaking of prodigies, yet mentions no con-
nection with omentum) : Juv. XIII. 117 “Τὰ
carbone tuo charta pia tura soluta Ponimus
et sectum vituli 1ecur albaque porci Omenta.’
Later, Macrobius uses it of periosteum, ‘ Ut
os secetur, omentum quod impositum est ossi
cruciatum, dum sectionem patitur, importat
(Sat VII. 9), and of the Meninges, ‘ cerebrum
...non suo sensu, sed vestitus sul, id est
omenti, hunc importat dolorem (ibid.).
In none of these passages is there the
slightest hint of connection with omen ; nor
even in Varro, who is not slow to connect
similarity of sound with similarity of origin ;
‘Indidein (ab ore) Omen, Ornamentum ;
alterum quod ex ore primum elatum est
Osmen dictum, alterum nunc cum praeposi-
tione dicitur volgo Ornamentum, quod sicut
157
olim Ornamentum scenici plerique dicunt.
Hine Oscines dicuntur apud augures quae
ore faciunt auspicium’ (De LZ. Lat. VI. 76).
The derivation from omen seems to me by
the above arguments sutliciently disproved :
but there remains the difficult task of
assigning a satisfactory origin to Omentum.
By modern etymologists several sugges-
tions have been made : (1) = augmentum 1.6.
fat: (2) contracted from ovimentum, as if
originally confused with the ovary : (3) con-
tracted from operimentum (Scaliger, 1619,
on Aristot. H.A. I. 136 says ‘Omentum,
valde concisa voce, fuit enim operimentum ’) :
so Scarabelli, who also suggests (4) akin to
Celtic ‘om’ = in, on, whence idea of ‘ inward
parts.’ ‘Omentum dal celt. ‘‘om,” lo stesso
che ‘am,’ sopra, intorno, tutto all’ intorno,
se pur non ὃ sincope di operimentum, coper-
tura’: (5) derived from same root as ὑμήν.
This last view has found many supporters ;
Vaniéek with Bauchfell and others would
refer these words to a root ,/au = to clothe,
or cover, seen in induo, exuviae ; while Byrne
in his Origin of Greek and Latin and Gothic
Roots (1888), tracing language to its primitive
phonetic origin, says that the sound ‘w’
conveys the idea of an enclosure, and a like
idea is conveyed by ‘m’: thus we get on the
one hand ὑμήν, o-mentum; on the other
μήνιγξ, mem-brana, o-men-tum. In any case
supporters of this view attribute to Omen-
tum the primary meaning ‘veil’ or ‘cover.’
In favour of this is to be said that Aristotle
several times actually uses ὑμήν in the de-
scription of Omentum, eg. ἔστι δὲ τὸ μὲν
ἐπίπλοον ὑμὴν τοῖς μὲν στέαρ ἔχουσι στεατώδης
τοῖς δὲ πιμελὴν πιμελώδης (P.A. IV. 3. 1),
ἔστι δὲ (τὸ ἐπίπλοον) τὴν φύσιν ὑμήν (Π.4.1.
136), ἔστι δὲ καὶ τὸ ἐπίπλοον ὑμήν (H.A. 11].
174), οἵ. Athenaeus p. 107 ὁ ἐπίπλους εἴρηται
ἐπὶ τοῦ λίπους καὶ τοῦ ὑμένος.
The Omentum does present exactly the
appearance of a curtain or veil concealing
most of the small intestine when the ab-
domen is opened ; and this sense of ‘covering’
would account for the later use of the word
= periosteum, and meninges.
The choice, I think, les between this last
Jau or ,/u, and a root ,/op, seen in ὀπός,
and possibly in ops, probably closely allied
with the root of πίων, opimus, ete.
It is evident that in most of the places
where it occurs, especially in descriptions of
sacrifices, the main idea of the passage is
‘fat,’ e.g. in those quoted above, ‘Omentum
inflamma pingue liquefaciens’ (Catull.), ‘Tot
tibi cum in flammis iunicum omenta liques-
cant’ (Persius), ‘tremat omento popa venter’
(Persius), ‘ Omenta pingues viscerum obtend-
108 THE CLASSICAL REVIEW.
unt sinus’ (Seneca), ‘ ventriculus atque intes-
tina pingui ac tenui omento integuntur’
(Plin.). The corresponding Greek word ἐπί
πλοὸν is constantly associated with words
meaning ‘ fat,’ e.g. in Aristotle quoted above
(P.A. 1ν. 8. 1) and elsewhere ; so the κνίσσα of
the Homeric sacrifice isexplained by scholiasts
as ἐπίπλοον, 6.5. κατά τε κνίσσῃ ἐκάλυψαν
Δίπτυχα ποιήσαντες, Whereon scholiast, Διπλ-
σαντες τὴν κνίσσαν, νῦν δὲ λέγει τόν ἐπίπλουν,
and Hesychius, δίπτυχα: δύο ἐπιβολὰς ἔχοντα
ἢ διπλοῦν, ὥστε τῷ μὲν ὑπεστρῶσθαι τὸν ἐπί-
πλοον τῷ δὲ ἐπιβεβλῆσθαι.
Further, the Semitic representatives of
Omentum, eg. Arabic and Syriac ‘Tharb’
and ‘Tharba,’ are said to mean ‘fat,’ and
indeed this latter sense is the only one that
survived in the form ‘Zirbo,’ used in Sar-
dinia.
Lastly, perhaps the most obvious character-
istic of the gastrocolic fold of the peri-
toneum, the Omentum, is the quantity of
adipose tissue present. Quain speaks of it as
sometimes ‘loaded with fat.’ On the whole,
therefore, I am in favour of the root /op, to
which perhaps should also be referred
omasum (said however by Philoxenus, Gloss.,
to be Gallic), and the very late word omin-
ada, = ‘bosse, tumeur, abscés’ (Maigne
and Migne, Lew.).
Gero. F. STILL.
HADLEY’S HIPPOLYTUS.
Hippolytus. With Introduction and Notes
by W.S8. Hapiey. 12mo, pp. 146. Cam-
bridge University Press. 28.
THis excellent edition of the Hippolytus
begins with a refreshing protest against the
‘ravages’ committed upon ancient texts by
critics whose ‘vastness of learning’ is not
‘directed by infallibility of taste.” ‘The
authority of the MSS.,’ this editor thinks,
‘is surely of some weight.’ He remarks
(p. 124) that the Hippolytus has suffered
especially, in the way of wholesale excision
and alteration, from ‘the will-o’-the-wisp-like
influence’ exercised on the ingenuity of
editors by the first edition (the ‘Im. καλυπ-
τόμενος). Accordingly, we have here a text,
by no means unduly conservative, of which
every page gives proof of care, judgment,
and taste. The well and even brilliantly
written introduction contains some admir-
able remarks upon ‘ Euripides in relation to
his times,’ and a comparison of the treat-
ment of the subject by Euripides, Seneca,
and Racine: and there is a good excursus on
Orphism. The notes are full, not too full:
there is enough, and not too much, of the
views of other editors, and in their treat-
ment no carping but no lack of indepen-
dence: the style is clear and vigorous: the
scholarship, the critical acumen, the literary
appreciation leave nothing to be desired.
Difficulties—of which there are many in the
Hippolytus—are never avoided. The ac-
curacy of the printing, too, deserves praise.
One mistake is worth noticing, πιθοῦ Soph.
Tr. 387 (in note on 491) for πεύθουι (The
quotation does not seem apposite.)—Lines
29—33 are a well-known crux. Mr. Had-
ley, retaining the rest, rejects 32, 33, which
indeed seem indefensible ; for, to say nothing
of the identification of the temple with the
Ἱππολύτειον, Phaedra could never have so pro-
claimed her love: and he accounts ingeniously
for their insertion, though it may be doubt-
ed whether the interpolator intended τὸ
λοιπὸν to mean ‘ posterity ’ and to be the sub-
ject of évopalev.—43. ‘ public’ and ‘ private’
should be dropped: πολέμιος is a declared
foe, ἐχθρὸς an enemy in feeling who may or
may not be πολέμιος. 78. The justification
of ‘the bold figure of Αἰδὼς tending and
keeping ever fresh the beauty of the meadow,
as a maiden’s purity is ever reflected in her
face,’ is excellent. For εἴληχεν ‘has taken
them for her own,’ Pind. Ol. 1, 53, ἀκέρδεια
λέλογχε Kaxaydpos might have been quoted.
87. ‘round the goal’ is not satisfactory.
κάμπτειν with τέλος, and the like, seems to
have acquired a secondary meaning, to ‘run
home’ along the second limb of the δίαυλος.
It is well explained by Mr. Bayfield on Jon,
82. 88. The point of the yap clause 1168,
perhaps, less in the antithesis of ἄναξ and
δεσπότης than in its serving as an apology
for the offer of advice in the next line. 93,
99. The fault of the logic is more apparent
than real. ‘We all hate σεμνότης (pride):
and so do gods. Do not with your pride af-
front the proud goddess Aphrodite: leave
pride to her.’ 104-7. Is the transposition
necessary? The same doubt arises on 330,
332, and on 1045-8. 135-8. ‘The pure form
of bread’ can hardly be right. If Orphie
THE CLASSICAL REVIEW.
language is admissible here at all, fasting
might very well be described as ‘keeping
the body pure from bread.’ And κατ᾽ dp-
Bpoctov στόματος may be defended, if Adu.
ἀκτᾶς δέμας ἁγνὸν ἴσχειν is regarded as the
opposite of σῖτον δέχεσθαι : on the principle
laid down by Wickham on Hor. Od. 1, 35,
28, ‘ferre iugum pariter dolosi,’ that ‘/erre
depends not on the whole idea of dolosi, but
on the positive attribute denied in it.’ 141
sq. Mr. Hadley reads with Wecklein οὐ γάρ,
and in 145 ἢ οὐκ (where Wecklein οὐδ᾽.)
finding a trace of this in the end of φοιταλέου,
a variant for φοιτᾷς in the previous line.
He rests his interpretation (‘you are not
frenzied : are you not rather wasting away
because of Dictynna?’) on a not very ob-
vious antithesis between φοιτᾷς and τρύχει.
It seems better to read ἢ οὐκ and οὐ yap...
ὀρείας aS a question: ov = nonne in both
cases. ‘Are you not frenzied by Pan, etc. ἢ
or are you not plagued by Dictynna?’ 169,
Mr. Hadley reads σὺν θεοῖς ἐφοίτα for MSS.
σὺν θεοῖσι φοιτᾷ: a great improvement.
290. The editor deals tenderly with the ap-
parent confusion in a former edition of
εἱπόμην With εἶπον. 306. “ μή not οὐ μεθ., as
putting a contingency.’ Rather, because of
the imperative ἴσθι. A note might have
been added on ἴσθι προδοῦσα ‘ know that you
will have betrayed’: as if θανουμένη were
understood, ‘ know that you will die having
betrayed.’ 322. γὰρ with a question is
surely conjunctive, asking for an explanation
of what has been said. 375. ἄλλων for
ἄλλως seems unnecessary, and does not go
well with θνητῶν. ἄλλως ἐφρόντισ᾽ is very
suitable for the ‘old trains of vague thought,’
when her own conduct was not yet in ques-
tion. 383 sq. Mr. Hadley rejects 384-7, and
transposes 388-390. ‘The pannus,’ he says,
“is not even purpureus.’ But is it therefore
not Euripidean? The λεσχαὶ καὶ σχολή,
which are among the distracting pleasures
of life, seem to be ‘ prating and philosophy’
in lieu of action, making cowards of men : and
there is an aidws which restrains from decisive
action when decisive action is needed—the
temper of Hamlet—a scrupulousness akin to
that which restrains from crime, virtuous per-
haps or at least amiable in its origin, but its
indulgence is one of the forms of ἡδονή and its
results are ruinous. 388. The suggestion that
the quasi-auxiliary use of τυγχάνω is in its ori-
gin ‘ironical’ is ingenious : a simpler sugges-
tion is that it arose from the desire to express
coincidence of this and that. 441. MSS. οὔτ᾽
(or ox) dpa γ᾽ οὐ δεῖ τοῖς ἐρῶσι τῶν πέλας.
Valeckenaer οὐ τἄρα λύε. Mr. Hadley conj.
οὐκ Gp’ ἀγὼν δὴ τοῖς ἐρῶσι viv μέγας, | ὅσοι τε
159
μέλλουσ᾽, εἰ θανεῖν αὐτοὺς γρεών ; but the line
has a patched look, and the meaning re-
quired is ‘so much the worse for lovers’ ;
whereas ἀγὼν μέγας ‘the issue is a great one’
is used of a critical decision still pending, or
a critical situation on which much depends,
and the point of the phrase lies in the uncer-
tainty—e.g. Hipp. 496 viv δ᾽ ἀγὼν μέγας,
σῶσαι βίον σόν (‘it is touch-and-go with your
life’): Med. 235 κἀν τῷδ᾽ ἀγὼν μέγιστος, ἢ
κακὸν λαβεῖν ἢ χρηστόν (‘to get a bad or a
good one, all turns on that’): Republi. 608
Β μέγας 6 ἀγών...τὸ χρηστὸν ἢ κακὸν γενέσθαι:
cf. Bacch. 975, Rhes. 195. If Valkenaer’s
reading is adopted, τοῖς ἐρῶσι τῶν πέλας Must
mean ‘ your neighbours who are in love,’ not
‘those who are in love with their neighbours’
(to which Mr. Hadley justly objects). 467.
Mr. Hadley retains ἀκριβώσειαν, but prefers
ἀκριβώσαις av. 491. He conjectures ὡς τάχος
δὲ πειστέον. But διιστέον seems satisfactory :
‘we must know one way or the other,
whether he will or he won't.’ Certainly the
stop after τἀνδρὸς is right. The ‘ coarseness,’
objected to by Weil, is of the essence of the
speech. 504. The objection to ‘my heart
has been well prepared by love,” i.e. is
already prone to yield’ is that it takes no
account of μὲν and dé. Bothe’s od for εὖ
may be right: this gives antithesis: but if
not, the dat. ἔρωτι is difficult, because am-
biguous—possible however, as a dat. of
reference (which the antithesis seems to re-
quire), not of agent. ‘Love finds my soul
prepared’ (or ‘ finds me mistress of my soul’ :
whichever meaning of the verb is preferred).
Or we might say that the idea of prepared-
ness (or self-mastery) implies resistance.
‘Now I am proof against passion ; but, if
you gloze the shame, all my good resolu-
tions will be swept away by the tide.’ 506.
ἀναλωθήσομαι εἰς ay very well be a refine-
ment upon the colloquial φθείρεσθαι εἰς, as
the Schol. explains it: διαφθαρήσομαι καὶ
ἐμπέσω (πεσοῦμαι) εἰς ὃ φεύγω, “1 shall be
swept, with a loss of all that I prize and all
that I believe, along the tide of passion.’
07. εἴ τοι δοκεῖ σοι, ‘If that is your way of
looking at it.’ Rather ‘If you will have it
so. The nurse pretends to give way, de-
ceiving Phaedra. At first (477) she talks of
ἐπῳδαί: ‘you are in for the malady: trust
to me, and I will bring you safely through
it’; meaning to persuade Hippolytus.
When Phaedra protests, she says plainly,
‘We must tell Hippolytus the truth’
(490-7). Phaedra still protesting—‘ Pro-
ceed no further in thismatter’—now she says,
‘ Well, if it must be so—a pity that you went
wrong, but we will do what we can’ [dev-
100
τέρα γὰρ ἡ χάρις, viz. of what I shall now
propose : ‘this is the next best thing I can
do for you:’ not, as Mr. Hadley, ‘ gratitude
is a secondary consideration’|; ‘you shall
be cured by charms: there shall be no dis-
grace —and the charms (don’t be afraid of
them) will not overset your reason. But
the remedy requires that 1 should procure
some word or token from Hippolytus, to
make it effectual.’ ‘How am I to take it?’
asks Phaedra. To which the only answer is,
‘Leave all to me.’ ‘I am afraid you will
tell him.’ And again, ‘Leave all to me.’
Certainly Phaedra is weak, and neither
trusts nor restrains the nurse. But it is
too much to say that she ‘suggests, under
guise of warning, the very course the nurse
has resolved upon.’ She says (weakly
enough no doubt), ‘I am afraid you will
do after all what just now you threatened.’—
This analysis agrees with Mr. Hadley’s
view, except that it is rather more favour-
able to Phaedra, and that it retains 513-5,
which Mr. Hadley rejects, not quite consistent-
ly with the claim that his reading of the scene
‘involves no surgery.’ 670, 1. Mr. Hadley’s
ψόγου for λόγου is an improvement : other-
wise he follows the MSS. 678. βίῳ (‘at
the expense of life’) for βίου is questionable :
βίου seems to require δυσεκπέραντον. 715, 6.
These lines have been much rewritten. Mr.
Hadley retains 716 (with δή τι) unaltered :
but in 715 conjectures and prints ἕν δὲ
πρόσθ᾽ εἰποῦσ᾽ ἐρῶ, which surely is weak.
But προστρέπουσ᾽ (which the Schol. explains
ζητοῦσα καὶ ἐξερευνῶσα) can hardly stand.
Perhaps ἐπιστραφεῖσ᾽ (‘upon reflection’), or
even ἐπιστρέφουσ᾽ : conceivably προσστρέ-
φουσ᾽΄. Cf. the use οὗ προσβάλλειν (τὸν νοῦν)»,
Soph. Zr. 580, 844. 724. Mr. Hadley ex-
plains, ‘Do you at any rate (unlike the
nurse) give me good advice.’ But the καί,
and the repeated εὖ, link the words ironically
with the speech of the Chorus, εὔφημος ἴσθι.
‘Yes, and don’t spare your good advice : but
T must die.’ 921, 2. Mr. Hadley thinks that
these lines refer to Socrates, ‘ who then, in
the streets of Athens, was trying to convince
men of ignorance, hoping by conviction to
THE CLASSICAL REVIEW.
drive them into virtue.’ 983. He adopts
Herwerden’s conjecture ἕύντασις, remarking
that évoracis would rather mean ‘sullen-
ness.’ 988. ἔχει μοῖραν, not exactly ‘has a
due share of influence,’ but nwmerwm aliquem
habet, ‘is of some account,’ ‘ claims some
consideration ’ (lit. ‘ fills some room,’ οἵ. ἐν
μεγάλῃ, οὐδεμιᾷ μοίρᾳ, etc.) : ‘this also (to
please the select audience) is of some value :
for it is just what the other sort cannot do’
(it is those who cannot please the wise who
please the crowd). 1186. As to ‘potential
optative without av,’ in Prom. 291 ὅτῳ is in-
terrog. (see A. Sidgwick, Agamemnon, ap-
pendix on ‘Remote Deliberative’): and in
Bacch. 747 one MS. has σὲ ξυνάψαι and the
other (pr. man.) od Evvawar. But with this
θᾶσσον ἢ λέγοι τις οἵ. Agam. 552 τὰ μέν τις εὖ
λέξειεν εὐπετῶς ἔχειν (where however Sidg-
wick argues for ἂν λέξειεν) : Aves. 880 ὥσπερ
εἴποι τις τόπος : Androm. 929 ὡς εἴποι τις:
Iph. A. 1210 οὐδεὶς πρὸς τάδ᾽ ἀντείποι βροτῶν.
1297. ‘Yet will I not make thy path easy.’
Rather, as Paley, ‘I shall gain nothing by
it (towards restoring the dead to life).’
1349. χρησμοῖς well explained ‘the answer
of heaven to the curse of his father’: the
god bestowed (ἔχρησε) what Theseus unjustly
asked. 1416. ‘ Let be, for otherwise it will
be the worse for you’: ἄτιμοι ‘unhonoured,
i.e. without effect’: 1.6. Cypris will pursue
you, not vainly, with her hatred even
when you are dead. But δέμας will not
bear this meaning: see Mr. Hadley’s own
note on 131. Artemis says ‘let be: for
even when you are dead Cypris shall not
wreak her rage upon you unpunished’:
meaning, ‘when you are dead she shall be
punished for having wreaked her rage upon
you.” For ἄτιμοι -- ἀτιμώρητοι see L. and 8.
It is echoed by τιμωρήσομαι (1422), and
τιμὰς (1424). Then 1419 (which is not very
like 1454) may remain. 1441. More readers
will agree with Mr. Hadley, who thinks this
‘the most beautiful line in a most beautiful
scene,’ than with Prof. Mahaffy, to whom it
seems ‘impossible, from such an artist as
Euripides.’
R. WHITELAW.
FLAGG’S IPHIGENIA IN TAURIS.
Euripides’ Iphigenia among the Tauwrians.
Edited by Isaac Frace. Ginn & Co.
1889. Square 8vo. Pp. 197. 4s. The
Text separately, 10d.
We have already in this country a very good
school edition of the /phigenia in Tauris ;
but Prof. Flagg’s book is also a good one
and will be found useful even here, designed
as it is for younger students than Mr.
England’s, students who have hitherto read
no Greek verse but Homer and are brought
to a standstill by each fresh crasis or
Doricism. There is therefore none of that
THE CLASSICAL REVIEW.
critical matter which occupies Mr. England's
footnotes, but only an incomplete list of
deviations from MS. authority at the end of
the volume. Mr. Flagg has constructed his
text with common sense, but yet I think
that for boys Mr. England’s is the better.
For scholars of course neither of the two
is satisfactory, inasmuch as both are full of
conjectures which have no probability at
all; but for boys the matter stands other-
wise. We cannot always give them what
Euripides wrote, for the simple reason that
we have it not: the next best thing then is
to give them what he might have written,
grammar and good sense; and this is what
Mr. England’s freely amended text attempts
to do. Mr. Flagg’s more conservative re-
cension no doubt comes quite as near as
Mr. England’s to what Euripides wrote, but
it contains much more which Euripides not
only did not write but could not; and the
learner’s conception of Greek is impaired
accordingly. The notes, if we set those
aside which explain the various inexplicable
things retained in the text, suit their pur-
pose well: they are terse yet not niggardly :
especially admirable is the way in which
they keep the reader’s eye on the progress of
the action, and bring to his notice those
strokes of art which a schoolboy intent on
construing is sure to miss. Here and there
the taste of an islander is offended by a
style which breathes the ampler ether and
diviner air of America, but otherwise it is
only details that challenge demur. 250.
“τοῦ ξυζύγου τοῦ ἕένου: of the stranger who
was his mate. 'The constr. seems to be like
ὃ ἀνὴρ 6 ἀγαθός, treating ξένου as an adjec-
tive.’ But the translation treats €évov as a
substantive, and requires for parallel 6
ἀγαθὸς 6 ἀνήρ. 266. ἄκροισι δακτύλοισι πορθ-
μεύων ἴχνος, ‘ferrying his track on tiptoe.’
300. ὥσθ᾽ αἱματηρὸν πέλαγος ἐξανθεῖν ἁλός, “50
that the briny deep bloomed forth with
gore’: we need not encourage boys to
translate in this style. 740. Amusing notes
have been written on this passage, but none
soamusing as Mr. Flagg’s. Iphigenia wishes
Pylades to swear that he will carry her letter
to Argos ; Orestes absurdly stipulates that she
on her part shall take oath to send Pylades
safely out of the Chersonese, to which she
very naturally replies, ‘Why, how could he
carry my letter unless I did?’ Other com-
mentators have remarked on the ineptitude
of Orestes; Mr. Flagg on the contrary
admires the noteworthy cleverness of Iphi-
genia, and lest we should think that such
acumen transcends probability he explains
it by the fact that she had not learnt to
NO. XXXII. VOL. Iv.
161
read and write. 790. “ἐμπεδώσομεν : will
make good ; cf. ἔμπεδον, ν. 758.’ The student
who does compare y. 758 and finds that
ἔμπεδον is there translated binding will be
puzzled to reconcile the two notes. 902.
Priestess and victim have been revealed to
one another as sister and brother, and forget
their peril in transports of emotion ; Pylades
now interposes to remind them where they
stand. Mr. Flagg hits off the situation with
Columbian vivacity by saying that Pylades
‘calls time.’ 1152. ‘The text,’ says Mr.
Flagg, ‘is a more than Terpsichorean maze.’
It may be, but he should not therefore adopt
from J. H. H. Schmidt an emendation which,
after violently altering the MSS., produces
gibberish. When it comes to introducing
an οὐκέτι of which the text has no vestige,
and then translating this οὐκέτι as if it were
οὔπω, one prefers Terpsichore. 1415. The
text gives one punctuation, the note assumes
another. 1418. φόνον τὸν Αὐλίδι ἀμνημόνευ-
tov θεᾷ προδοῦσ᾽ ἁλίσκεται : Supposing that
this meant anything, by what imaginable
jugglery could it be made to mean, ‘is found
guilty of betraying the goddess’ trust in
that forgotten murderous deed at Aulis’ 1
The introduction, which covers fifty pages,
is a full and interesting account of the
legend, plot, artistic structure, and metres
of the play. Mr. Flagg calls this ‘nega-
tively considered, the most faultless of
Euripides’ extant tragedies,’ and thinks that
‘there remains not another one that is
marred by so few of those grave lapses from
dramatic propriety and universal good taste
to which the poet’s mind was subject.’ Even
negatively considered I should have thought
the Hippolytus by far the most faultless
tragedy of Euripides, if not indeed the most
faultless of all tragedies except the Antigone
alone: what lapses mar it, apart from a
certain artificiality in the altercation of
Theseus with the hero, I do not know;
assuredly none to compare with the see-saw
of divine intervention in the ἔξοδος of our
play. Mr. Flagg defends this machinery as
the only way to rescue the chorus, which is
one of those excuses which Aristotle calls
ridiculous ; the poet, as he says, should take
care from the outset not to construct his
play in suck a manner. And when Mr.
Flagg says that ‘the modern reader cannot
adequately reproduce the feelings stirred by
this final scene in the Athenian spectator’s
breast,’ this is to arraign Euripides, not to
defend him. It means that he wrote for an
age and not for all time; he defaced his
drama that he might gladden the eyes of
the vulgar with the resplendent stage-pro-
M
162
perties of their beloved goddess: a trap to
catch applause which does not differ in kind
from the traditional sentiment, always wel-
come to the gallery of our own theatres,
that the man who lays his hand upon a
woman, save in the way of kindness, is
unworthy of the name of a British sailor.
On p. 36, where Mr. Flagg speaks of the
increased employment of a resolved arsis in
Euripides’ later plays, the terms, ‘less care
in the finishing,’ ‘deterioration,’ ‘laxity,’
‘degeneracy,’ ‘looseness,’ give a wrong im-
THE CLASSICAL REVIEW.
pression. We are free to prefer the earlier
practice ; but the change proceeds from no
negligence or failure in craftsmanship. The
Bacchae, but for the injuries of time, is
hardly less exquisite in finish than the Medea
itself ; only it is written on other principles.
The text is printed clearly and, so far as
I have observed, accurately, except for the
common mistake ya, χοῦς, χἅτερον, χὑμῖν,
rough breathing for coronis; θοὐμόφυλον
however is given correctly.
A. E. Housman.
ULLRICH’S STUDIA TIBULLIANA.
Studia Tibulliana, de libri secundi editione,
scripsit Ricoarp Uuuricu. (Berolini, R.
Weber. 1889. pp. 86.) 1 Mk. 80.
THE author deals principally with the vexed
questions of the chronology of ‘Tibullus’
poems. Ov. 77. i. 463, 464 ‘legiturque
Tibullus et placet et iam te principe notus
erat’ is to be interpreted to mean that Tibul-
lus published the first book in 27 B.c. (nine
months after Octavian had the title of Au-
gustus conferred upon him). The second
book, in spite of its brevity, i